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 2007-07-31

00:01:10pondlifeAh, that's only file removal.
00:01:47pixelmapondlife: Slasheri told me "auto-update" works without dircache too - it is just slow
00:02:10pondlifeThe deleted files check is #ifdeffed out
00:02:52pixelmaI remember having asked him once
00:03:17 Quit Xerion (Read error: 104 (Connection reset by peer))
00:03:42Soul-Slaye1Why is it the diff file I've made in Windows has a different layout to diff files I've made in Linux before nwo
00:04:02pixelmafrom the wiki: If "automatic update" has been set to on, it automatically does everything that "update now" does at every boot. However, without dircache (General Settings -> System -> Disk -> Directory Cache) enabled, auto update skips the check for deleted files because that would take very long without dircache. "Update now" includes that check whether dircache has been enabled or not. - whatever that means
00:04:38pondlifeSounds about right.
00:05:01pondlifeAlthough tagcache.c also has other checks for dircache that I don't quite get.
00:05:44pondlifeIt should probably treat flash players like they do have a dircache.
00:06:11amiconnSoul-Slaye1: The charge and discharge percentages are different on purpose
00:06:31pondlifeAnyway, time for sleep.
00:06:38pixelmapondlife: ah, didn't see your second line - was already typing the reply to your first post
00:06:52JdGordonpondlife: i dont reemember that patch, but if it works, go for it, im heading to work in a few min
00:07:10amiconnThey are captured from real discharge / charge cycles on a couple of targets. The discharge and the charge curves always look different
00:07:16pondlifeJdGordon: I've no idea if it works, I don't have the appropriate number of bits available to test it :)
00:07:26JdGordonpondlife: oh, no, that has been commited
00:07:37pondlifeSo I can close that bug?
00:07:38Soul-Slaye1amiconn: Ok. I can't understand why, but fair enough
00:07:45JdGordoni dont think the aim of that was to fix that bug though
00:08:08midgeyany opinions on FS #7383
00:08:18JdGordonamiconn: can you test that patch pondlife linked to so it can be closed? (64bit sim bug)
00:08:22amiconnWell, a current flow requires a voltage difference (unless the total resistance is zero - which only applies to superconductors)
00:08:40pondlifeJdGordon: If n1s == nls then maybe he can tell us....
00:08:51amiconnAnd the chemistry of rechargeables also adds to that
00:08:52Soul-Slaye1midgey: Surely disk mode is handled by the apple OS
00:09:04midgeyit was closed once and reopened, but the freeze occurs in diskmode
00:09:14Lloreanmidgey: Why is 7383 even opened? Who reopened it?
00:09:21JdGordonis there an n1s in the house?
00:09:37pondlifeJdGordon: I've put a message on FS.
00:09:38LloreanOr better yet, why did Zagor reopen it?
00:10:02midgeyLlorean: no idea. it was you who closed it originally
00:10:15midgeyi think it should be closed since its not our issue
00:10:49 Quit JdGordon ("Konversation terminated!")
00:11:05pondlifeAnyways, g'night all.
00:11:08 Quit pondlife ("zzzz")
00:11:47 Quit stalwart1 (Connection timed out)
00:12:06salty-horsefirmware/bidi.c has several c++ style comments (used to comment out debug/temp code)
00:12:27Lloreanmidgey: I definitely think it should be closed, for the same reason
00:14:32 Join DerPapst [0] (
00:14:53 Quit davina (Remote closed the connection)
00:15:40 Join Stalwart [0] (
00:16:28 Quit Stalwart (Nick collision from services.)
00:16:39 Join Stalwart [0] (
00:17:25*DerPapst just came here to point out that there is already a bootloader available for 1G and 2G iPods that is able to load rockbox. This loader is even older than the 1G/2G rockbox port. It's (the probably by all rockbox devs hated bootloader called) Loader 2! :P
00:18:28Bagderloader 2 something beta, isn't it?
00:18:55amiconnWell, on the g1...g3 the release version might even work
00:19:10amiconnWe're not running dual core on PP5002 yet
00:20:05amiconnAlthough, I prefer bootloaders without extensive menus...
00:20:17DerPapstLoader 2 is pretty stable on all my iPods (which are 2) and i haven't heard anythin bad on the iPL forums about that either
00:20:50amiconnAnd using rockbox (and ipl from what I read) is still somewhat cumbersome on 1st/2nd gen, due to the lack of shutdown
00:20:58DerPapsti load rockbox with loader 2 and it works fine...
00:22:52DerPapstbut when amiconn introduces dual core support on PP5002 that wont work anymore. and i bet amiconn isn't willing to fix loader 2 :'(
00:23:02 Join rotator [0] (n=e@rockbox/developer/rotator)
00:23:10 Join Soap_ [0] (n=Soap@rockbox/staff/soap)
00:24:24LloreanWhatever happened to fixing the things that need to be fixed for 2nd core playback?
00:24:27Bagderimho, someone should add the menu thing to our bootloader in an #ifdef thing
00:24:43*preglow reappears
00:25:28amiconnBagder: Why would you want a menu in the bootlader?
00:25:30Soap_Llorean, regarding your pedantic ;) correction to the question of how much money scorche won with my betting...How did mr.weathers get a rockbox.ipod file in his root directory if he is correct in that he first installed Rockbox only 1 month ago?
00:25:37amiconnpreglow: wb :)
00:25:40Bagderamiconn: I don't, but clearly lots of people do
00:26:03Bagderand telling them to go ipl for the loader only isn't an ideal situation if you ask me
00:26:04LloreanSoap_: Maybe he first installed an unsupported build (jbuild?) that hasn't been updated for a long enough time to include that change?
00:26:25LloreanBagder: I agree, solely because I'd rather have them using something we can support.
00:26:42*preglow reads 600 rockbox mails
00:27:18*amiconn thinks the only thing the bootloader is lacking is speed
00:27:22pixelmaBagder: JdGordon mentioned that he started something like that just a few days ago - something with holding a button at boot to get to the menu *shrug*
00:27:26Soap_good thinking Llorean. Once again exposing a covert assumption on my part.
00:28:13DerPapstpixelma: but he didn't come far and gave up ;)
00:28:33DerPapstso noone is working on it courrently
00:28:33pixelmaah yes...
00:28:44*amiconn isn't sure what's the best way to restructure the IRQ handler
00:28:47 Quit Thundercloud (Read error: 113 (No route to host))
00:28:55amiconnIt now needs to dispatch various GPIO interrupts
00:31:01 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
00:31:47linuxstbBagder: I would prefer to see effort made to get rolo working and loading the Apple firmware - we could then just add a "Load Apple Firmware" option to the main menu.
00:32:25Lloreanlinuxstb: Or the system menu?
00:32:26linuxstbOr even add that option via a reboot with some magic in IRAM...
00:33:04LloreanMaking Rockbox itself handle the job of multiboot wouldn't be bad either, but I think you might still have a hard time convincing people not to use iPLL2 under those circumstances.
00:33:50linuxstbI haven't used it for a while, but I wonder what the difference is in booting IPLL2 compared to Rockbox...
00:34:30amiconnlinuxstb: There are other magic cookies than 'diskmode\0' and 'hotstuff\0' defined in the ipod flash rom...
00:34:32linuxstbRockbox Utility could even start writing rockbox.ipod to the firmware partition, getting rid of the bootloader completely...
00:34:37Soap_Since once an iPod is in Apple Firmware it tends to stay in Apple Firmware, baring long-term disuse, the rolo or IRAM option doesn't sound so unreasonable for those who like IPL
00:34:54amiconnI wonder if anyone has tried them
00:35:00linuxstbamiconn: I know there's a retailos check
00:35:07 Join Thundercloud [0] (
00:35:49linuxstbIt's used at least by the diskmode "app" - writing that to IRAM before launching the diskmode app will cause it to reboot after disconnecting.
00:35:57*amiconn needs to test the available magic cookies on 1st/2nd gen
00:36:03 Quit HellDragon ("reboot")
00:36:22amiconnThere are less than on the later ipods
00:36:36DerPapstThere is one thing the rockbox bootloader or rockbox itsself can't handle readring bootlin iPodLinx which Loader 2 can. Rockbox can't pass argument to the kernel or getLoader2Args.
00:37:02DerPapstthat would be the reason why i wouldn't use the rockbox bootloader or rockbox to boot iPL
00:37:11linuxstbThen write a patch...
00:37:27DerPapstwhich won't get acceped anyways...
00:37:31LloreanA "Boot iPL" plugin would. :-P
00:37:44LloreanIt could parse a text file containing the same options as work with the iPL Loader 2
00:37:53preglowamiconn: you have a 2g ipod now?
00:37:59DerPapstesp. not for the bootloader because that would require a menu
00:38:13amiconnpreglow: Yes, and soon a 1st gen as well
00:38:17linuxstbpreglow: Welcome back...
00:38:17LloreanDerPapst: As I said, make it run from Rockbox using a plugin that parses the same config that Loader2 uses.
00:38:43LloreanThen it'd be mostly compatible, easily launched, and have little to no reason not to accept it.
00:38:57 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
00:38:58donutman25how do i dump sansa firmware with the bootloader?
00:39:21preglowlinuxstb: thanks
00:39:29amiconnpreglow: I had to RE battery adc access as that is something the ipl people never dived into... and I have it working on 2nd gen :)
00:39:30DerPapstas you probably know all i've done so far is redifining some buttons for plugins.... so i doubt i'm somewhat of able to code that plugin.
00:39:31preglowdone with vacations now
00:39:54amiconnAnd we definitely need sleep mode as the PP can't be powered off on the 1st/2nd gen
00:39:56DerPapstand as long i can't ROLO the kernel in rockbox there is even no point in doing that.
00:39:58 Join chrisjs169 [0] (
00:40:08preglowamiconn: yeah, i was about to ask if you've aquired a taste for disassembly :P
00:40:23linuxstbDerPapst: I've rolo'd the IPL kernel in the past.
00:40:47amiconnIt seems that the OF actually calls the ROM functions to access the ADC, as the main firmware doesn't seem to have functions for that
00:40:51DerPapstinfact... i'm not even 100% how to do that in rockbox...
00:41:04LloreanDerPapst: Last time I checked, RoLoing iPL actually worked.
00:41:13DerPapstwell then...
00:41:15LloreanAt least until it got to the point where it realized I didn't have an ext3 partition.
00:41:24linuxstbAt the moment Rolo only works with .ipod files - so run scramble on the kernel, and just select the .ipod file in the file browser.
00:41:25amiconnSo if we are able to run rockbox from flash one day, it might be problematic to get the OF to run...
00:41:36*DerPapst starts his preferred coding editor...
00:41:44preglowanyone heard anything from saratoga? seems not much has happened since i left
00:41:56Lloreanamiconn: I think if we're running from Flash we don't need to be too concerned about dual boot.
00:42:15amiconnRoLo... on archos that's possible
00:42:40 Quit ender` (Read error: 104 (Connection reset by peer))
00:42:47linuxstbpreglow: He's made a few commits I think. But there's not much left to do in terms of features - seeking is the main thing left, plus fixing problems with the occasional troublesome file.
00:43:18amiconnlinuxstb: Isn't there this special mode for low-bitrate files that's still missing? lsp?
00:43:19Lloreanamiconn: Well I'll admit it'd be nice to be able to rolo the OF, I don't think it should be seen as a necessity when running from flash. For example, we can't yet on H100
00:43:26 Quit jgarvey ("Leaving")
00:43:51linuxstbamiconn: Maybe - I've lost track. I know he's made some low-bitrate fixes of some kind though.
00:44:04preglowlinuxstb: yeah, i asked him to look into seeking before i left. heard anything about progress?
00:44:08DerPapstLlorean: but there is one problem though... but i don' know the Loader 2 internals at all. The string (arg) that getLoader2Args is reading from seemingly out of nowhere have to be stored somewhere. Maybe in ram.. and i don't know what rockbox would say when i start randlomly writing stuff in the ram.
00:44:14Soap_don't want the ipods to surpass the HXxxx's now do you ;)
00:44:24preglowhe's working on lsp, afaik
00:44:28preglowstill not working 100%
00:44:41preglowbut i just got back from england, haven't spoken to him for a week
00:44:47LloreanDerPapst: Well, it's just a suggestion. As I don't use iPL at all, I must admit not being particularly invested. ;)
00:44:52*Llorean goes to dinner.
00:45:09*petur goes to bed
00:45:15 Quit petur ("Zzzzzz")
00:45:52DerPapstI mean even if i would want to write that plugin... if i can't store the args anywhere the whole thing simply doesn't work.
00:46:03linuxstbpreglow: He mentioned about week ago that he was going to investigate seeking, but haven't heard anything about it since...
00:47:09linuxstbDerPapst: I would imagine that you would need to add a new (small) function to the Rockbox core which actually handles the call to the main rolo() function and passes the arguments. Your plugin would then call that function.
00:48:07linuxstbAs it's an ipod-specific feature and all ipods have at least 32MB of RAM, I can't see the few bytes it would take being the issue. The main one IMO is if anyone would want to use it...
00:49:13DerPapsti would if that is actually the only solution to have rockbox, iPL and the OF triple booting.
00:49:26 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
00:49:33DerPapst*once dualcore support is introduced
00:49:49linuxstbWhat's the problem with dual-core and loader2? It should work fine.
00:50:44 Join advcomp2019 [0] (n=advcomp2@
00:51:03DerPapstafaiu introducing dual core support on pp5003 breaks loading rockbox with loader 2
00:51:37 Part advcomp2019 ("Leaving")
00:51:47linuxstbI don't think that's the case.
00:52:08DerPapstwell then all is good :)
00:52:31amiconnlinuxstb: So why does an older bootloader not work for dualcore rockbox on PP502x then?
00:52:57 Quit chrisjs169 ("CGI:IRC (Ping timeout)")
00:53:59DerPapstlinuxstb submitted a patch to loader 2
00:54:01linuxstbOlder bootloaders don't, but current versions should work fine.
00:54:19 Quit obo ("bye")
00:55:10DerPapstlinuxstb: by lloking at getLoader2Arg.c i couln't even find out where it gets the argumenst from ^^
00:55:10amiconnlinuxstb: Yes, but why does what makes the older rockbox bootloaders not work not apply to ipl2?
00:55:31linuxstbBecause ipl2 was fixed
00:55:47linuxstb(as DerPapst said, I gave them a patch which was committed)
00:56:04amiconnYes, but that patch isn't in the release version yet, is it?
00:56:15*amiconn never tried loader2
00:56:28DerPapstit is.
00:56:34DerPapstlong long times ago
00:56:40amiconnWell, perhaps I unknowingly did when I tried ipl for a very short time on my mini g2
00:56:50DerPapstinfact.. i made the release version of Loader 2.5
00:56:55Soap_dare I ask the SVN command to check out an old revision?
00:57:05linuxstbDerPapst: The Loader_2 wiki page has been updated? For a long time after my patch was committed, that still showed 2.4 as the current release.
00:57:15bluebrotherSoap_: svn co -r <releasenumber>
00:57:20DerPapstyes it has been updated.
00:57:53linuxstbIt says "18 May 2006" ...
00:57:56DerPapstthere were even a Loader 2.5d6 version available with your patch. But the wiki hasn't been updated
00:58:05DerPapsti forgot the date then ;)
00:58:38linuxstbDoes that also include the change to look for rockbox.ipod in .rockbox/ ?
00:59:02DerPapstand i think this "last updated" row is stupid. since mediawiki adds the date automatically at the bottom of the page.
00:59:20Soap_bluebrother, thank you but "svn co -r 13990" from my /rockbox directory on ubuntu gives me a "not enough arguments provided" error.
00:59:29DerPapstcourtc added half of my patch to svn
00:59:52DerPapstthe other half was bumping the laoder version from 2.5d6 to 2.5
01:00:13midgeysoap: i thought the command was svn up -r 13990
01:00:13DerPapstso the loader version in svn seems older than the release one ^^
01:00:36bluebrotherSoap_: sorry, you need to svn up -r<number>
01:00:45Soap_aye - that seems to do the trick.
01:01:07Soap_I was going to spend my time here in the hotel building older nano builds to help with the diagnosis of the nano problems.
01:01:07bluebrothersvn co -r<number> works too, but you need to specify the svn server as this is a checkout
01:01:13DerPapstbut in loader 2 is still the hack included to get the hw version... ;)
01:01:22Soap_does anyone, Llorean, know what versions should be built for testing?
01:01:47Soap_ahh, that makes sense bluebrother
01:01:56linuxstbDerPapst: You spotted my commit to remove the need for that?
01:02:03linuxstb(commit to Rockbox)
01:02:54linuxstbAnd I guess it will also need updating to load Rockbox for the 1g and 2g.
01:03:11linuxstb(it won't recognise the header in the .ipod files)
01:03:34DerPapstah.. haven't thought of that.
01:03:59*DerPapst wonders where they are...
01:04:12*bluebrother wonders if someone would be interested in creating some fancy side graphics for rbutil's windows
01:04:28DerPapstfound them...
01:05:07DerPapstsimply adding them to static char* rockboxIDs[]? can it really be that easy?
01:06:15DerPapstnot quiet i guess...
01:08:48 Quit salty-horse ("Leaving")
01:09:14*DerPapst stops talking to himself and goes to sleep instead.
01:09:29DerPapstgood night everyone ;)
01:09:33 Quit DerPapst ("So Long And Thanks For All The Fish!")
01:10:43 Join hannesd_ [0] (
01:12:06linuxstbYou'll also need to initialise "sum" to 19 - the model number for the ipod1g/2g. And it's a single ID - "1g2g"
01:12:39***Saving seen data "./dancer.seen"
01:13:09 Part toffe82
01:13:10 Quit linuxstb ("CGI:IRC")
01:14:30 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
01:15:32linuxstbsoap: Did you say you have a candidate for the standard codec test track?
01:20:22 Quit jhulst (Read error: 113 (No route to host))
01:20:41Soap_yes, it is at home while I am on the road
01:21:05linuxstbSo I assume you can't access it?
01:21:42linuxstbI think it's about time we stopped searching for the perfect track and just picked something...
01:21:53Soap_it is a meta-track. made of a classical, a rock, a hip-hop, and an electronic piece from's Open-source CC-SA archve.
01:22:14Soap_when the wife gets home from school I'll see if I can't talk her through FTPing it for me.
01:22:35scorcherdp/ssh is handy for that =)
01:22:50linuxstbDon't worry about it - we've waited this long... (unless you're on the road for 6 months)
01:22:52Soap_it fails the > 200 kbps LAME -V2 test, but it looks decent on a frequency graph
01:23:17Soap_I should be home friday night or Satuday morning depending on how up to the drive I feel Friday.
01:23:33 Join advcomp2019 [0] (n=advcomp2@
01:23:42Soap_scorche, my home computer is 450 MHz, RDP is really pushing it. ;)
01:24:07linuxstbYou could run sshd in Cygwin...
01:24:23scorcheSoap_: it may take an hour while it refreshes, but you would at least get it done
01:24:31scorcheand ssh doesnt take much
01:24:53Soap_ssh I could do I guess
01:25:05 Join BobShield [0] (
01:25:12amiconnLooks like my interrupt drived ADC scan is working :D
01:25:45linuxstbDid you say that it's just used for battery monitoring?
01:25:57 Join liberty____ [0] (n=pocketir@
01:26:16 Quit hannesd (Read error: 110 (Connection timed out))
01:26:17 Nick hannesd_ is now known as hannesd (
01:26:25linuxstbAnd that's currently twice a second?
01:26:35amiconnBut a conversion can take several milliseconds, so the interrupt driven read yields while waiting for the result
01:29:30amiconnThe 1st gen adc is way faster (well, according to my understanding of the flash rom disassembly
01:30:10amiconnIt uses delay loops in the range of a few cycles (up to 80 - I guess that's at 24MHz so I used 320 for now)
01:31:16amiconnThe 2nd gen uses delay loops of 2500 loop cycles, plus it is supposed to wait for the ready bit, but that's broken if I am not totally wrong with my interpretation
01:32:15amiconnJust delay loops weren't enough to get a stable readout - I had to implement waiting for the ready bit, which was what made me think about using interrupts for this wait
01:35:51 Part n1s
01:36:07 Quit Soul-Slaye1 (Read error: 104 (Connection reset by peer))
01:36:48 Join jhulst [0] (
01:37:44 Quit midgey ()
01:41:19 Join Soul-Slayer [0] (n=jonno@
01:41:26amiconnjhMikeS: What's the deal with priority_yield() vs. yield()?
01:41:35amiconnI.e. which one should I use...
01:42:25 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
01:44:28Soap_is exploring exactly what svn revision is causing problems on the Nano with Apple FW version 1.3.1 wanted?
01:46:35 Join kubiix [0] (
01:47:53 Quit bluebrother ("leaving")
01:52:06 Quit Robin_0800 (Read error: 104 (Connection reset by peer))
01:52:14 Quit secleinteer (Connection timed out)
01:52:24 Join secleinteer [0] (
01:55:59 Quit XavierGr ("One firmware to rule them all!")
01:56:55 Quit liberty____ (Read error: 110 (Connection timed out))
02:04:55 Quit kubiixaka (Read error: 110 (Connection timed out))
02:05:48 Quit Siltaar (Remote closed the connection)
02:07:59 Join joshin_ [0] (n=joshin@VDSL-130-13-11-18.PHNX.QWEST.NET)
02:08:55 Quit linuxstb ("CGI:IRC")
02:09:39 Quit joshin (Nick collision from services.)
02:09:57 Nick joshin_ is now known as joshin (n=joshin@unaffiliated/joshin)
02:15:08 Join maddler [0] (
02:15:13 Quit hannesd ("Client suicide")
02:15:15maddlerhi all! :D
02:15:30*maddler just discovered rockbox! :D
02:16:00maddlerbeside I'm unable to mount my e260 on my Ubuntu 7.04 box :)
02:16:15maddlereverything looks fine on the device...
02:16:45maddlerand in /var/log/messages: Jul 31 02:14:42 aubergine kernel: [24451.928000] usb 5-4: new high speed USB device using ehci_hcd and address <nn>
02:16:55maddlerwhere "<nn>" keeps being incremented...
02:17:02maddlerbot nothing more...
02:17:05maddlerany clue?
02:18:26scorchemaddler: you need to use the OF (original firmware) for USB
02:18:44Soul-SlayerUSB support has yet to be implemented to Rockbox
02:18:57Soul-SlayerYou need to go into the old firmware and use USB from that until it's added ^^
02:19:11maddlersad... :(
02:19:21maddlerrockbox looks _GREAT_!
02:19:39maddlerso... I'm supposed to go back and forth with sansapatcher?
02:19:53Soul-SlayerI'm under the impression dual boot is implemented...
02:19:54Soap_no, there is dual-boot
02:20:19Soul-SlayerHold down the Left button while it boots up
02:20:23Soul-SlayerShould boot you into the OF
02:20:52maddlerlemme try
02:21:06maddlersince sansapatcher won't work... ;)
02:21:32Soul-SlayerYou'll only need sansapatcher again now if a new bootloader is released, or if you decide to remove Rockbox entirely.
02:22:16maddlerbtw... sansapatcher is actually complaining about no E200 connected or not in UMS mode...
02:22:19maddleris that fine?
02:22:55maddlerI know... I should read the FAQ... :)
02:23:47Soul-SlayerIt will unless you're in the OF, I believe
02:24:01Soul-Slayer*Doesn't own a sansa but assumes it won't find it if it's not connected* ^^
02:26:09maddlerhehehe... it's connected of course...
02:26:32maddlerok... sansapatcher works once in OF :)
02:27:31maddlera BIG thank you to whoever is involved in rockbox development!!!
02:28:36Soul-SlayerEnjoy ^^.
02:30:45maddlerI'd be happy to contribute somehow...
02:31:01maddlerbut considering I'm everything but a programmer... :)
02:31:01Soul-SlayerKnow C? :P
02:31:08maddlerdunno... some beer? :DDDDD
02:31:15maddleror some real italian pizza?
02:31:32Soul-SlayerWell if you wish to donate beer I'm sure nobody will complain
02:31:51maddlerI bet!
02:32:40Soap_room 21, Davis Motel, under the strip club, Salem Ohio. Large with black olives and extra garlic
02:32:49 Join sharpe [0] (
02:32:53Soap_I'll be sure to share with those who actually do the work here.
02:33:23Soul-SlayerNaturally :p.
02:33:32scorchethere is a paypal button on the rockbox main page...we use it to fund our meetings together and for hardware as needed, if you are really anxious o contribute =)
02:37:44Soul-SlayerDo manuals rebuild whenever they are updated?
02:38:02scorchethey are built daily iirc
02:38:07maddlerSoap_: are you a developer? can you prove it? :D
02:38:13Soul-SlayerAh thats okay.
02:38:17scorchemaddler: he isnt =)
02:38:41maddlerSoap_: are you cheating?!
02:38:52maddlerwell... you coult try to bribe me... :D
02:38:54scorchehe is a significant figure though....someone who is involved in the project =)
02:39:07Soul-SlayerThats the role I wish to achieve some day :P
02:39:13Soap_I said I'd pass some pizza on to those who do the actual work...
02:39:16Soul-SlayerI made it into the credits *proud*
02:39:31scorcheyou can identify people involved with the project by looking at their hostmask
02:40:00maddlerscorche: ok... :)
02:40:06scorcheeg, soap's says n=Soap@rockbox/staff/soap
02:40:12maddler /foreach user /who
02:40:21Soap_which is shorthand for "worthless bitch"
02:40:33Soap_well, except in SodEater's case
02:40:49Soap_and, well, everyone else with that hostmask but Soap
02:41:03scorchenaw...meant for manual writers, support people, "experts", people who are generally involved with the project
02:41:22sharpeeh, i don't have a hostmask like that :)
02:41:28scorchethere is contributor, staff, developer, and administrator
02:41:38scorchesharpe: you have been gone for a long time...
02:41:50scorcheand before i "took over" irc
02:41:56Soul-SlayerHow can I earn me a contributor hostmask :P?
02:42:08sharpecontribute seems reasonable
02:42:23maddlerbtw... if you need an e260 beta tester... drop me a line...
02:42:38scorcherequirements for the contributor is to have your name in the credits
02:42:52Soul-SlayerMy name is in the credits, I want my cookie :(
02:42:54scorchei am thining about adding donator as well for paypal...
02:44:07LloreanSoap_: We know which revisions are causing it.
02:44:10sharpehonestly i think i'd be better off at working on backend stuff (webserver-wise, or rockbox even), but as far as i know we don't need any of that, do we?
02:44:13Lloreanre the Nano problem
02:44:36Soul-Slayermaddler: You can try going through the Sansa bug reports and verifying them, if they've given steps to reproduce
02:44:51Soap_sub $1000 = donator. $1000-$5000 = GoldenDonator, >$5000 = PlatinumClubDonator with access to the VIP lounge and champagne room?
02:44:57Soap_really Llorean?
02:45:18Soap_Dang, I just built 20 some odd nano versions for nothing? ;)
02:45:33 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
02:45:49LloreanSoap_: It's one of amiconn's clock speed changes, though I can't remember which off the top of my head.
02:46:03LloreanI think there's even a reasonable certainty on the diagnosis for why it causes a problem
02:47:17Soap_I have 1.3, not 1.3.1 on my Nano. I'll be honest and say I haven't upgraded my Nano for a while. Is 1.3 of value for testing once a fix comes up, or is 1.3.1 the only one with issues?
02:47:26Soap_iTunes does not offer me an upgrade past 1.3
02:48:01maddlerSoul-Slayer: ok... I'll give a look... :D
02:48:15maddlerdamn... this stuff _really_ rocks!!!
02:48:21maddlerand I mean *REALLY*!!
02:48:52sharpemaddler: what rocks?
02:49:01maddlerrockbox! what else?! :D
02:49:10sharpeoh, right, haha.
02:49:20LloreanSoap_: I'm pretty sure both 1.3 and 1.3.1 should show the problems on Nano. If you check the thread in the forum I've posted two test builds, the first should work the second should fail, and any build after like... 14004 or so should fail. Maybe 14003.
02:49:23maddlerI'm literally amazed with all the funtions/options/freedom it has!!!
02:50:15LloreanSoap_: Assuming memory is serving me correctly on those revision numbers
02:50:15maddlerI bought e260 today... and I was kinda happy with OF...
02:50:30maddlerbut rockbox makes it the definitive player!!!
02:51:02Lloreanmaddler: Only if you haven't been spoiled on the more mature Rockbox players. :-P
02:51:07 Join midgey34 [0] (
02:51:11 Join scorche` [0] (i=Blah@rockbox/administrator/scorche)
02:51:11 Quit midgey34 (Read error: 104 (Connection reset by peer))
02:52:38maddlerLlorean: ...?
02:52:51maddlerLlorean: which player? :)
02:53:00maddlertell me!
02:53:39Soul-SlayerAny of the older ones
02:53:55sharpemaddler: i think he means the ones that have been in development for much longer than the newer players
02:54:07Lloreanmaddler: H100, H300, X5, M5, or Toshiba Gigabeat.
02:54:10Soul-SlayerThe iriver H100/H300's are highly commended, and the gigabeat imo
02:54:20maddleroh... yup..
02:54:20 Quit BobShield (Read error: 104 (Connection reset by peer))
02:54:21LloreanThe Archoses are also very mature, but due to their processors are not quite as feature rich.
02:55:09Soul-SlayerAnd the fact you could build a house out of them tends to put people off.
02:55:21 Quit scorche` (Read error: 104 (Connection reset by peer))
02:55:22sharpethat'd be an interesting project
02:55:44 Quit saratoga ("CGI:IRC")
02:55:51LloreanSoul-Slayer: I actually like players with some momentum to them when the time comes to explain things to certain users. :-P
02:56:19Soul-SlayerHeh ;)
02:56:39pixelmathe Ondio is quite small...
02:58:34pixelmabut maybe I interpreted "build a house of it" the wrong way - wouldn't make me wonder today
02:59:03sharpecomparable to bricks is what came to mind for me
02:59:05Lloreanpixelma: I'm betting he was only referring to the JBRv1 / Player
02:59:41pixelmawell, you only said "Archoses"...
03:00:10LloreanWell yeah, *I* meant all of 'em. But a lot of people forget the Ondios (including me from time to time, I'm sad to say)
03:01:41pixelmaand he only said "them" - but anyways, I just wanted to drop a good word for my player before I'm off ;)
03:02:21Soul-SlayerI did mean all of them yes, you could still build a kennel or something out of Ondios :p
03:02:35Soul-SlayerAdmittedly it wouldn't be very water tight with it's... Unique... Shape
03:02:39 Join Chronon [0] (
03:03:51ChrononGreetings. I just signed up for a Twiki account and I was wondering if someone could give me write permission.
03:04:55 Join piroko [0] (
03:05:10 Part pixelma
03:06:03 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
03:07:02 Quit scorche (Connection timed out)
03:07:47 Join Llorea1 [0] (n=llorean@
03:07:58 Quit Llorean (Nick collision from services.)
03:07:59 Nick Llorea1 is now known as Llorean (n=llorean@
03:08:32 Quit YouCeyE (Remote closed the connection)
03:12:43***Saving seen data "./dancer.seen"
03:13:48 Part piroko
03:13:53Soul-SlayerAnyone know the average battery life on the Sansa?
03:14:40sharpeaccording to the wiki, maybe about 11 hours?
03:14:45 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
03:15:01advcomp2019i get about 18 to 21 hours with my sansa e280r
03:15:05LloreanSome people have claimed 14-15, and with frequency scaling now, it might even reach that. Depends an awful lot on the conditions though
03:15:13Lloreanadvcomp2019: With Rockbox?
03:15:15 Quit Soap_ (Read error: 110 (Connection timed out))
03:15:22 Join aliask [0] (
03:15:34advcomp2019i cant rockbox my r series
03:15:46LloreanAh, didn't see the R
03:15:54Soul-SlayerNot if it's an R. Can someone take a look at ? I'm not sure if it should be closed or not, there's little information given
03:17:06LloreanSoul-Slayer: Leave it for now, see if he provides something useful. Since he's in active conversation still.
03:17:20Soul-SlayerWhich does make a nice change =$
03:17:23advcomp2019i am waiting for rockbox for the r series
03:17:33Lloreanadvcomp2019: If everyone with an R just waits, it'll never happen.
03:17:34Soul-Slayeradvcomp2019: Could be waiting a long time
03:17:55Soul-SlayerAlright it didn't take two of us to emphasise that point, my bad ^^.
03:18:40Soul-Slayer But theres some info on it incase you're wondering why.
03:18:49maddlerany plan for USB support on e260? :)
03:18:55Battousaii get about 11 hours on my sansa, yeah
03:19:03maddlerkinda frustrating to have to reboot to OF in order to upload files...
03:19:04Soul-Slayermaddler: I doubt it's just being left off :p
03:19:18Soul-SlayerMost likely buried in a todo list somewhere
03:19:24maddlerjust curious... :)
03:19:29advcomp2019one thing i do not like with the OF is the voice recorder recording in wav format
03:19:44aliaskmaddler: Someone is working on it right now
03:20:00Soul-SlayerThere we go ^^
03:20:20 Quit HellDragon (Client Quit)
03:20:38ChrononI just registered for a Twiki account under the user name: "JustinHannigan". Can someone give me write permission?
03:20:45maddlerok... more beer on the go! :D
03:20:50aliaskChronon: Just a sec
03:21:00Chrononok, aliask.
03:22:06 Quit Slowking_Man (Client Quit)
03:22:15aliaskChronon: Should be good to go
03:22:54Chrononaliask: Thank you!
03:23:07 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
03:23:33 Join HellDragon_ [0] (n=Nocebo@unaffiliated/helldragon)
03:23:49 Quit HellDragon (Nick collision from services.)
03:25:56 Nick HellDragon_ is now known as HellDragon (n=Nocebo@unaffiliated/helldragon)
03:26:23advcomp2019i know a lot of people with the r series that wants rockbox... i tell them to wait or try to get the non r series
03:26:38Soul-SlayerOr you could ask them to help port it ^^
03:28:01Battousaito help entertain yourself as you pass the time, you could ask sandisk to do it
03:28:26advcomp2019they are not programmer or smart enough to play do it tho
03:29:58saratogaadvcomp2019: basically, find someone with an R series to do it or accept that it will never happen
03:30:16saratogaalso, does anyone know if the battery bench issues on the Sansa were solved?
03:30:20gxtirockbox failed me on the drive home today
03:30:47sharpegxti: pray tell why
03:30:51gxtiit started skipping, then crashed, and did all kinds of nasty things ... of course, this was a month-old build because the latest ones won't even play :p
03:31:10sharpeoh noes!
03:31:25Soul-Slayergxti: Target :o?
03:31:36gxtiipod nano
03:32:11Soul-SlayerA flash based player started skipping, youch
03:33:06LloreanSoul-Slayer: There are IDE timing issues right now
03:33:16gxtiyep... one song it was playing fine, the next it was jumping forward rapidly, a bit of fiddling and it locked up, then dumped core :P
03:33:16Lloreangxti: So, the same kinds of symptoms?
03:33:20Soul-SlayerOn a build a month old?
03:33:25LloreanThere's no "dumped core"
03:33:28 Join AnthonyMa [0] (i=470ceb56@gateway/web/cgi-irc/
03:33:31LloreanPlease state clearly what happened.
03:33:35gxtiwell, it didn't dump core... it was some error to that effect
03:33:55LloreanThere's no error that includes the word "Core" as far as I know.
03:34:03advcomp2019i have a r series but i do not want to brick it
03:34:06LloreanThere are illegal instructions, data aborts, and several others.
03:34:07gxtii know. i don't remember what it said, but i think it had 'dump' in it
03:34:11gxtidata abort. that's it.
03:34:28gxti... somehow, data + abort turned into 'dump' in my mind
03:34:37LloreanIt may very well be the IDE timing issue. What you may want to do is boot the apple firmware just once, then boot the old build of Rockbox.
03:34:47LloreanAnd see if that keeps your iPod stable for a while
03:35:01gxtihmm, i just booted it up and it's playing again
03:35:17gxtiwhich is odd, because i reset it a few times earlier and it got worse
03:35:59gxtiit must have powered down completely at some point (it wouldn't stay off)
03:37:02 Join Nick_Brackley [0] (i=cb57494b@gateway/web/cgi-irc/
03:37:07 Join EbErT [0] (
03:37:10 Join mag [0] (
03:37:41AnthonyMahi guys. i just registered a wiki account to upload a theme i just created, but it seems i need permissions set first by someone. is there anyone on at the moment that can help me out?
03:38:10aliaskSure - what's your wiki name?
03:38:22magCan anyone send me a link to the EVILG build bootscreen bmp? He apparently released them for the latest build, but I have no idea where to find them.
03:38:52aliaskAnthonyMa: Should be done.
03:39:04AnthonyMathanks a lot
03:39:10AnthonyMamag: here's a link to the latest one
03:39:24AnthonyMa4th post from the top
03:41:08magdo you need to be registered to see links in posts on rockbox forums?
03:41:19 Join midgey34 [0] (
03:41:20 Join chrisjs169 [0] (
03:42:44sharpei don't believe so
03:42:57Soul-SlayerLinks, no, attachments, possibly
03:43:40sharpeyou definitely do for atachments...
03:43:47sharpeattachments, eh.
03:44:11 Quit Chronon ()
03:44:27LloreanThe attachment has been removed anyway
03:44:31LloreanI missed spotting it.
03:44:43LloreanI'm really, seriously tired of people renaming file extensions to bypass the limitations on attachments
03:45:06LloreanDo they honestly think it's going to simply be okay if they rename it, because the name is what I set the limitations in place for?
03:45:13LloreanYes, I'm horribly offended by the term "zip"
03:45:55Soul-SlayerI wouldn't have thought so, but apparently people do.
03:46:16Soul-SlayerWhich file types are allowed?
03:46:21magtysm anthonyma!! <3
03:47:20Lloreantxt for logs, pdf for datasheets (though I'm tempted to remove that since the wiki is there for that), jpg/gif/png for compressed screenshots, .c/.h for code snippets in discussion
03:47:52Llorean.doc because I was told some important information has been found only in .doc in the past, .bin for a reason I can't remember, and .cfg because we end up asking to see the contents of those too often
03:47:58LloreanBasically it's restricted to "fairly small files" too.
03:48:18LloreanI've run out of space in the avatar/attachment directory a few times as it is.
03:48:19Soul-SlayerI was just thinking, bin is a fairly random extension to allow
03:48:37Llorean.bin is what gets used for dumps of flash and such
03:49:15Soul-SlayerWell, logs can be put in pastebin, as can cfg's
03:49:28Soul-Slayerjpg, gif, png, plenty of image hosting sites around..
03:49:56LloreanYes, but in both of those cases, *we're* asking them for the file
03:50:01 Quit GodEater_ (Read error: 110 (Connection timed out))
03:50:02LloreanSo we should provide a location for them to put that file.
03:50:16Soul-SlayerTrue enough.. C and H files we wouldn't though, would we?
03:50:19LloreanAs a general rule, things are allowed because they're files we'll ask users for.
03:50:28LloreanYeah, I'm not entirely sure why .c and .h made the list
03:50:36LloreanI think they were on the list before I took over so I grandfathered 'em in.
03:50:44Soul-SlayerFair enough.
03:51:13Soul-SlayerHow much webspace is available to
03:51:33LloreanI dunno
03:51:38LloreanIt's donated space, and I haven't been told.
03:51:56LloreanThe attachment directory was 10mb, I doubled it to 20 at one point, and nobody's complained.
03:52:00LloreanBut I don't want to expand beyond that.
03:52:24Soul-SlayerNo, well, the only file type that sounds like it could potentially horde that space is PDF
03:52:25 Quit jhMikeS (Nick collision from services.)
03:52:31 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
03:53:07Soul-SlayerAlthough I can't be sure on the size of bin files as I've never needed them.
03:53:28LloreanDepends on what it's dumped from.
03:56:26magcan anyone who has a 5g iPOD and who uses the EVILG build help me change the bootscreen?
03:56:27 Quit Nick_Brackley ("CGI:IRC (EOF)")
03:57:56Lloreanmag: Questions about unsupported builds should be asked in their thread
03:59:07 Quit Thundercloud (Remote closed the connection)
04:00:41Soul-SlayerAck, I said I'd test pondlifes patch... Can't remember the number... *Runs to the logs*
04:01:36Soul-SlayerPanic over. :P
04:02:21 Part safetydan
04:02:21 Quit maffe (Remote closed the connection)
04:05:11 Quit thegeek (Read error: 104 (Connection reset by peer))
04:05:15 Join thegeek_ [0] (
04:06:00maddlerok... time to hit the bed... 4am here...
04:06:05maddlerlater dudes!
04:12:20 Quit dandin1 ()
04:18:55EbErTim looking at toshiba gigabeats on ebay. Can anyone tell me the main cause for them to just "not power up"? seeing a couple of those
04:19:30 Part mag
04:19:43aliaskI don't know that there's a "main cause", but if anyone would know it'd be toffe82 when he's online
04:20:19Soul-SlayerEbErT: Have they detailed anything else?
04:20:54EbErTthere are like 4 i've seen, "wont power up" is the only answer
04:21:04Soul-SlayerCan you link me to one?
04:21:10 Quit thegeek_ (Read error: 113 (No route to host))
04:22:36EbErThold on
04:23:53EbErTthat's one i hadn't seen yet
04:23:54 Join DogBoy [0] (
04:23:57EbErTthere are many it seems
04:27:07Soul-SlayerWell, for starters you wouldn't want that as it comes without accessories (AC charger is proprietary and, although not necessary, nice to have)... I guess it depends how adventurous you are... It gives no real information as you said, I've never heard of a case of one 'just not powering up', possibly a bad battery, which are easy enough to replace if you can find one designed for it... It's still a risk though
04:27:40DogBoywhere do I look to see what codecs are supported
04:27:54Soul-SlayerName a codec, it's probably supported.
04:28:20Soul-SlayerBut it'll be detailed in the wiki and more than likely the manual too.
04:29:11 Join thegeek [0] (
04:29:12DogBoythank you
04:30:11DogBoyah so no real audio then eh
04:30:31saratogawhich real audio did you want?
04:31:02DogBoynot sure
04:31:13saratogai think theres some interest in cook, and of course aac will work already if you put it in mp4
04:31:39DogBoyhow can I tell which kind the file uses?
04:31:54Soul-SlayerWhat extension are your files?
04:32:45DogBoyAUDIO: 22050 Hz, 1 ch, s16le, 20.7 kbit/5.86% (ratio: 2583->44100)
04:32:45DogBoySelected audio codec: [ffcook] afm: ffmpeg (FFmpeg COOK audio decoder)
04:33:38Soul-SlayerCook is not yet implemented, so they won't work, no.
04:34:32DogBoyoh well
04:34:44DogBoyI can use real player on my palm pilot
04:35:13 Quit chrisjs169 ("CGI:IRC")
04:37:56 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:38:23 Join miepchen^schlaf [0] (
04:46:35 Join kubiixaka [0] (
04:47:28 Join tedr0ck [0] (
04:56:44 Quit ShadowXP (Read error: 104 (Connection reset by peer))
04:57:12 Join ShadowXP [0] (
05:00:17 Quit tedrock (Read error: 113 (No route to host))
05:05:29 Quit kubiix (Read error: 110 (Connection timed out))
05:12:44***Saving seen data "./dancer.seen"
05:16:40 Join maffe [0] (
05:24:18 Join joshin_ [0] (n=joshin@VDSL-130-13-10-32.PHNX.QWEST.NET)
05:25:51 Quit joshin (Connection timed out)
05:26:08 Nick joshin_ is now known as joshin (n=joshin@unaffiliated/joshin)
05:26:59 Quit rotator ()
05:29:49 Quit Rondom (Nick collision from services.)
05:30:05 Join Rondom [0] (
05:41:40 Join helloworld1 [0] (n=pocketir@
05:45:36aliaskI'm still having problems with that stack overflow in the gigabeat s bootloader. I'm actualling thinking there may be some problem in the linking of the binary, could that be a cause of a stack overflow?
05:47:23 Quit helloworld1 (Remote closed the connection)
05:54:02jhMikeSaliask: It's the svn one right? I haven't had a single problem with it after fixing the tick task issue.
05:59:56aliaskjhMikeS: This is for the Gigabeat S
06:00:20 Join Veralis [0] (
06:01:14 Quit sharpe ("Leaving")
06:02:24VeralisCan anybody help me with cygwin and compiling rockbox? Or can anyone do it for me if i send you the patched source code?
06:05:10LloreanVeralis: What problem are you having.
06:05:51VeralisWhen I go to compile, it says "cc1: Invalid option 'long-calls'"
06:06:03VeralisRight after "Header parsed"
06:09:05 Join erifneer1 [0] (
06:10:06LloreanHow did you set up cygwin?
06:10:14jhMikeSaliask: right :) misread that.
06:10:19VeralisAnybody willing to help?
06:10:45jhMikeSVeralis: out of date gcc version?
06:10:45aliaskjhMikeS: But perhaps this is related - how was the tick timer causing stack overflows?
06:11:25jhMikeSaliask: It wasn't. It was causing panics when the backlight driver tried adding a tick task. kernel_init hadn't yet been called.
06:11:27VeralisIm not sure...
06:11:35VeralisI've never done this kind of stuff before...
06:11:37erifneer1jw, but with the tracker clean up, are patches more likely to be added?
06:12:03jhMikeSveralis: which build?
06:12:32VeralisNot sure?
06:12:33LloreanVeralis: How did you set up cygwin? What method did you use to install the cross-compilers
06:12:37jhMikeSwhich device are you making for?
06:12:53VeralisThe ones on the rockbox sites, and the Gigabeat.
06:13:28VeralisThe automatic method, and
06:13:33jhMikeSthe gigabeat is giving that?
06:13:44VeralisNo, cygwin when I try to compile.
06:14:21LloreanVeralis: Type arm-elf-gcc −−version
06:14:42erifneer1could someone give me a tip to get my patch (7111) to be included
06:15:00jhMikeSis long-calls even a valid non-arm option?
06:15:04LloreanVeralis: You did something wrong, that should be 4.0.3
06:15:16LloreanjhMikeS: Gigabeat is ARM
06:15:27VeralisUm, anyone want to compile it for me?
06:15:27jhMikeSHe said he was compiling cygwin.
06:15:43jhMikeSor that's how I misinterpreted it
06:15:57LloreanVeralis: You're better off figuring things out.
06:16:01LloreanOr maybe try the VMWare solution
06:16:21LloreanOtherwise any time Rockbox gets a new feature, or important bug fix, you'll be asking the same question
06:17:00erifneer1so no love for the podcast patch?
06:18:41 Quit Soul-Slayer ("Leaving.")
06:18:57 Quit saratoga ("CGI:IRC (Ping timeout)")
06:21:15 Quit Veralis ()
06:27:17 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
06:45:37 Part erifneer1
06:45:51 Quit lazka (Remote closed the connection)
06:51:33 Quit HellDragon (Client Quit)
06:54:05 Join ptw419 [0] (
06:58:17 Part Llorean
07:12:47***Saving seen data "./dancer.seen"
07:18:24 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
07:19:37 Quit donutman25 ("ChatZilla [Firefox]")
07:27:30 Quit HellDragon (Client Quit)
07:32:14 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
07:39:51 Part advcomp2019 ("Leaving")
07:53:01TiMiD[FD]I wonder why the clock plugin stores it's setings into arrays instead of a struct ...
07:53:25TiMiD[FD]operation unbloat
08:19:47 Quit gromit` (Remote closed the connection)
08:32:18 Join webguest52 [0] (i=c0231115@gateway/web/cgi-irc/
08:32:51 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
08:32:51 Quit Rob222241 (Read error: 104 (Connection reset by peer))
08:33:09 Join Rob2222 [0] (
08:36:55 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
08:39:20 Join ender` [0] (
08:39:43 Join Rob222241 [0] (
08:39:51 Quit EbErT ()
08:41:42 Quit joshin (Read error: 110 (Connection timed out))
08:46:24 Join GodEater_ [0] (n=bryan@rockbox/staff/GodEater)
08:48:34GodEater_morning folks
08:49:16JdGordonholy crap!
08:49:24aliaskHoly crap?!
08:49:31JdGordonuh, i mean morning!
08:49:39JdGordonstupid typos
08:50:35 Join pondlife [0] (
08:51:02GodEater_JdGordon: I took your point on board about my patch yesterday, and moved the .links file it creates/uses to the root directory. Since the plugin is also a viewer, you can launch it just by clicking on that file, which makes it a lot quicker to access.
08:51:49GodEater_although to keep amiconn and other "more root directory clutter" anti-fans happy - the file is only created if you choose to run the plugin, otherwise it's never there.
08:54:55 Join aarongoltz [0] (i=180f4ce8@gateway/web/cgi-irc/
08:56:25 Quit Rob2222 (Connection timed out)
08:56:54JdGordonGodEater_: ok, that sounds a bit more useful :)
08:57:26GodEater_indeed - I actually found myself using it yesterday, something I didn't think I'd do ;)
08:58:19aarongoltzI have an update on my issue
08:58:34JdGordonso is it actually going to get commited? or just sit in the tracker?
08:58:35aarongoltzI just tried installing the regular Ipod Video Rockbox firmware on my Ipod, and it worked fine
08:58:47aarongoltzWhen I tried the 64 meg version, it crashes the Ipod again
08:59:16GodEater_JdGordon: well, a) it needs testing more, b) I still need to write the feature that allows you to delete favourites from the file when you get bored of them, and c) I have no commit access
08:59:32JdGordonall valid points :)
09:01:11aarongoltzIs there something in the 64 meg version that accesses anything in my Ipod that might be broken, that the 32 meg version doesn't access?
09:01:22*GodEater_ is hoping to earn his developer badge doing this or similar work though
09:02:04GodEater_aarongoltz: yes, another 32MB of memory
09:02:10midgey34aarongoltz: what is the size of your hard drive in you iPod?
09:02:16aarongoltz60 gig
09:02:23aarongoltzMaybe some of my memory is bad
09:02:30aarongoltzand the 32 meg version doesn't access it
09:02:56GodEater_aarongoltz: did you acquire this ipod from new as a 60GB, or has someone upgraded the hd in it ?
09:03:00aarongoltzI didn't have problems using the Apple OS, but I'm going to test it now playing a video file
09:03:10aarongoltzIt was a new 60 gig Ipod
09:03:19 Quit HellDragon (Client Quit)
09:03:37aarongoltzAnd like I said, I was running 64 megs for a long time, since the 64 meg addition came about
09:03:39GodEater_interesting - maybe apple only upped the memory halfway through the production run
09:03:43aarongoltzWithout problems until a few days ago
09:03:52GodEater_that's very weird
09:03:53aarongoltzWell, I did have my logic board replaced
09:04:03aarongoltzCould they have replaced it with a 32 meg board?
09:04:09GodEater_very possibly
09:04:12aarongoltzBest Buy sent it out
09:04:39aarongoltzIs there any way for me to tell the amount of memory it currently has?
09:04:42GodEater_I've never actually heard of someone getting their ipod repaired before
09:04:46GodEater_usually you just get a new one
09:05:08aarongoltzYeah, Apple store is cool like that, but it turns out I didn't have a replacement plan at Best Buy, but a service plan
09:05:14GodEater_aarongoltz: I think the fact that the 64MB build crashes says it all
09:05:19aarongoltzThey don't just hand out mp3 players anymore, only Xbox's and PS3
09:05:45 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
09:05:51aarongoltzDoes anyone know roughly when 64 meg version came about?
09:06:10GodEater_about two weeks after the introduction of the 80GB build IIRC
09:06:16aarongoltzSince many months ago, after the logic board replacement, I went to reinstall Rockbox, and I think I ended up using an old version
09:06:27aarongoltzSince i had the same problem
09:06:44aarongoltzI left my Ipod in the freezer for 6 hours last night and it didn't break, unfortunately
09:07:34 Join petur [0] (n=petur@rockbox/developer/petur)
09:07:42aarongoltzWell then I guess I will try to research and see if there are 60 gig versions that aren't 64 megs, or if Best Buy just gave me a 32 meg logic board
09:09:11GodEater_wth is this leaving it in the freezer thing? Where's that come from suddenly ?
09:09:50aarongoltzA couple years ago, when I was a student in Denver, I took my 3G Ipod which has a messed up screen, stuck it in 3 ziplocks, then in the freezer for 4 hours
09:10:08aarongoltzWhen I took it out, it had a frowney face, and Apple sent me a new Ipod about 5 days later
09:10:26aarongoltzBut this morning when I took out my video, it was freezing cold, and went right on
09:10:35GodEater_"messed up screen" ?
09:10:51JdGordonwhat is everyones opinions on 2653 (with a bit of midification)? (choose the insert/queue option when selecting a file)
09:11:10aarongoltzTHere was corrosion to the screen cover so it was very hard to read without the backlight on
09:12:29aarongoltzI can't believe I've spent almost 20 hours trying to fix this all to discover that I might not have 64 megs of ram. I'm loading up the 32 meg version to confirm that it runs very well
09:12:49***Saving seen data "./dancer.seen"
09:13:21 Quit aarongoltz ("CGI:IRC (EOF)")
09:13:47 Join aarongoltz [0] (i=180f4ce8@gateway/web/cgi-irc/
09:15:20aarongoltzI don't know anything about the way Rockbox runs, but I'd like to throw out this theory
09:16:16aarongoltzWhen I would play a Flac file, it would crash after about 6-8 seconds. Mp3 would make it to a little bit longer. Once Rockbox tried loading the music into the Cache or ram, the second it tried to access the extra ram, it would crash
09:16:21aarongoltzIs that what was going on?
09:20:05GodEater_JdGordon: doesn't 2653 completely change the standard action? Or am I just not understanding it ?
09:20:29GodEater_ah I see
09:20:34linuxstbJdGordon: Regarding the click-to-insert patch, I seem to recall a lot of people objecting to it. Personally, I don't mind - it seems a waste to me for most targets having two buttons (select and right) doing the same thing.
09:21:18JdGordonwell llorean and bluebrother arnt here and they are the major objectors to configurable stuff... so i should fix and commit it before they get in :D
09:21:45GodEater_they'll never know!
09:22:48 Join davina [0] (
09:25:05GodEater_do long presses on a button generate a different ACTION event I can check for in my plugin ?
09:25:28JdGordonare you using the plug actions or direction button codes?
09:26:16*GodEater_ hopes this is the right answer =/
09:26:24JdGordonyeah, then umm... some do
09:26:32aarongoltzI saw that you posted a while back in the forums saying that the 64 meg's larger playback buffer allowed for increased battery life. If I only play Flac files, would I notice any substantial increase?
09:26:58JdGordonwhat you can go is call rb->get_Action_code(&some_variable) and it will dump the actual button press into some_variable so you can check for |BUTTON_REPEAT
09:27:58GodEater_I can't just check for ACTION_STD_CONTEXT ?
09:28:32GodEater_which appears (on ipod at least) to be BUTTON_SELECT|BUTTON_REPEAT
09:28:47GodEater_rather than the other action I'm checking for, which is ACTION_STD_OK
09:29:15*JdGordon not exactly sure what you want....
09:29:21GodEater_sure - sorry
09:29:47GodEater_currently I check for ACTION_STD_CANCEL, and ACTION_STD_OK
09:29:50JdGordonalso, remember, the button codes for the actions are not the same over all the targets... thats why it was introduced
09:30:02GodEater_that's why I'm sticking to the ACTION_ codes
09:30:13GodEater_so I don't have to worry about the targets
09:30:40GodEater_so I just want to add a check for ACTION_STD_CONTEXT
09:30:49 Quit aliask ("ChatZilla [Firefox]")
09:30:53GodEater_I think I have the answer to my question anyway ;)
09:31:12JdGordonnothing stopping you... all the ACTION_STD_* codes are almost always ok to check for
09:31:24JdGordonthe other ones depends on what context your "in"
09:31:34GodEater_this is for the "delete selected item" option that the plugin in missing currently
09:31:48GodEater_this would be in my list widget
09:32:47JdGordonyeah, just add a case for _CONTEXT or _MENU would maybe be better
09:33:24GodEater_perhaps you're right
09:38:13 Join gromit` [0] (
09:38:56 Quit gromit` (Remote closed the connection)
09:43:27 Join B4gder [0] (
09:44:48linuxstbGodEater: Is there a reason your plugin can only bookmark folders, rather than files? Also, reading the task comments, why would manually editing the file be likely to break it?
09:45:11GodEater_linuxstb: because my parsing code only looks for \n
09:45:15GodEater_not for \r\n
09:45:35GodEater_because I'm dead lazy ;)
09:45:36JdGordonyour not using rb->read_line?
09:45:48GodEater_JdGordon: no - it seemed wasteful for file with at most 10 lines
09:45:49linuxstbI think that's worth fixing. One of the nice things of text files for config files is that they can be edited in any text editor...
09:46:02GodEater_I shall mend it then
09:46:18linuxstbAlso things like trailing spaces and slashes...
09:46:21GodEater_also, folder favourites was all that was asked for
09:46:48JdGordonI havnt looked at the format, but you might like to use parse_settings() (or whetevr its called) if you have a format like the config files
09:46:56GodEater_I haven't
09:47:00JdGordonoh well
09:47:01GodEater_it's just a line with a path on it
09:47:53petur maybe the user wants to give it a fancy name (set it to the name of the dir by default)?
09:48:25*GodEater_ gets a pen
09:48:45linuxstbGodEater_: We're going to make you work for your developer badge ;)
09:49:16*GodEater_ doesn't object
09:49:42GodEater_you think having it mark things other than folders is worth it though linuxstb ? isn't that what bookmarks are for ?
09:50:23aarongoltzThanks for the help again tonight. I'm not going to let not having 64MB prevent me from being happy about having Rockbox running again
09:51:43JdGordonGodEater_: well, if your using set_)current_file() then there is no extra work for you
09:52:07linuxstbGodEater_: No, I think it's different to bookmarks. It just seemed a natural thing to do in your plugin - so in effect users can save both the folder and the position in the folder.
09:52:26GodEater_linuxstb: okay - well that change is trivial at least
09:52:28 Join bdgraue [0] (
09:53:00*GodEater_ hasn't looked at set_current_file()
09:53:15GodEater_actually yes he has
09:53:20JdGordonisnt that wjat your using to get the fileborwser into the right folder?
09:53:22GodEater_I think I need more coffee
09:53:49GodEater_does anyone think having the context of the file tree accessible by external code is a bad thing ?
09:53:49 Quit aarongoltz ("CGI:IRC (EOF)")
09:54:08GodEater_I had to add a function to the tree code to allow it to be changed after the tree had been initialised
09:54:25GodEater_specifically the dirfilter setting
09:54:38GodEater_usually you set that when you call rockbox_browse I think
09:54:45GodEater_and it's not then changed subsequently
09:54:46linuxstbAnother thing.... Why is there a limit of 10 bookmarks?
09:54:59GodEater_linuxstb: completely arbitrary
09:55:25GodEater_it seemed silly to allow lots more than that - otherwise you'd spend more time hunting through your favourites file than you would have if you'd just browsed the file tree
09:55:44linuxstbTrue, but that's up to the user.
09:56:01GodEater_....I guess so.
09:56:14linuxstbBut something like 50 could be better, and still easily fit within the Archos's 32KB of plugin RAM.
09:56:48*GodEater_ notes this to
09:59:01 Quit ptw419 ()
10:00:39amiconnjhMikeS: wow!
10:01:26 Join gandhi [0] (i=9088689c@gateway/web/cgi-irc/
10:02:14pondlifeGodEater_: Just catching up here, are you going to use rb->read_line()? I think you should...
10:02:28 Quit linuxstb ("CGI:IRC")
10:02:48GodEater_pondlife: well now that linuxstb thinks I should have more than 10 entries, it might make more sense
10:03:14GodEater_it seemed inefficient earlier when it was such a small file, I thought reading it all in one go and then parsing it in memory was better
10:03:18pondlifeBest to only have the newline parsing in one place.
10:03:28GodEater_I think I'm convinced
10:03:38 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
10:03:45GodEater_will also rename it to shortcuts, just for you
10:04:16pondlifeNot for me, for all the confused people who Llorean will have to explain it to... and for Llorean!
10:05:05pondlifeHmm, shouldn't the home page mention that the 1G/2G iPods are supported now they're in the build table?
10:05:18linuxstbNo - there's no bootloader.
10:05:23linuxstbAnd probably no manual...
10:05:23pondlifeGood point!
10:05:42linuxstbAnd no-one has tested it on the 1G...
10:06:20GodEater_why are there two columns ?
10:06:27linuxstbThe build was renamed
10:07:43pondlifeDoes anyone have a clue why voice use results in permanent CPU boost - ?
10:09:07linuxstbI can't remember anyone mentioning the reason - just that it happens
10:09:50pondlifeI'll try and work out which commit introduced that feature - it hasn't always happened.
10:11:20 Quit goffa (Read error: 110 (Connection timed out))
10:11:42 Join thegeek_ [0] (
10:12:47 Quit thegeek (Read error: 104 (Connection reset by peer))
10:14:04B4gderbug fs#7341 claims to contain a patch that fixes the problem, on ipod video
10:14:28B4gder"Freezes if booted with USB cable connected."
10:14:48B4gderthere's a close request filed for it, but I don't think we should close until the code is in svn or rejected
10:14:49*GodEater_ would love that fix :)
10:14:57 Join spiorf [0] (
10:15:11 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:16:02amiconnpondlife: 1st gen needs some different inits I will add when I get mine to test it on
10:16:54pixelmapondlife: I think jhMikeS said why the 100% boost with voice happen
10:17:04pondlifejhMikeS: ping
10:17:21pondlifeAlso does anyone think deserves to stay open?
10:17:29 Join nl3 [0] (
10:18:25pondlifeThere's not a lot to go on as a bug report.
10:18:27 Join My_Sic [0] (
10:18:30 Join obo [0] (n=obo@rockbox/developer/obo)
10:18:38B4gderI vote close
10:19:04pondlife'tis closed.
10:19:25B4gderafter all, people are very keen on requesting re-open when they feel strongly about a report
10:19:40*JdGordon was joking
10:19:54pondlifeJdGordon: I thought you may have been.
10:20:44 Part nl3
10:21:13 Join nl2 [0] (
10:22:27 Join n1s [0] (
10:22:37 Part nl2
10:23:16JdGordonbloody hell... all the flyspray emails in the last 2 days and we are still 977 open tasks!
10:23:38B4gderit's quite amazing
10:24:00pondlifeStill got 5 days left...
10:24:59linuxstbB4gder: Have you looked at the patch in FS #7341 ? It looks to me like a workaround which avoids the bug, rather than a fix.
10:25:23B4gdernope, I just noticed the close-request from the user
10:25:42B4gdersince I don't have a 5g nor am I familiar with the area really
10:26:37pixelmaJdGordon: about the one-click-insert patch: please keep the Ondio in mind. It doesn't have two buttons that do exactly the same as linuxstb said about the others...
10:26:44linuxstbIt's also not a 5g-specific bug. I get it at least on my ipod Color, and I think others have reported it on other targets.
10:27:05pondlifeI don't like the use of #ifdef in there, at least.
10:27:17pondlifeAnd there's no sign that anyone understands the problem.
10:27:28linuxstbpondlife: Exactly - the #ifdef rang alarm bells for me...
10:27:52JdGordonpixelma: right, so it would either not be used for ondio, or just use this instead of the default (it being customizable anyway)
10:28:09*pondlife wonders if the homepage's "Flyspray #" button will wear out...
10:28:13B4gderI just deny the close request
10:28:20linuxstbJdGordon: Does the click-to-insert patch work on folders as well?
10:28:35pondlifePut a note on to say that we need to understand what's going wrong?
10:28:44pondlife(to B4gder)
10:29:00B4gderthat's basically what I said in my reason for rejecting the close
10:29:07JdGordonlinuxstb: I didnt really look at the patch carefully (I'm not sure I like the way he did it), but a proper implementation would
10:29:26jhMikeSpondlife: yo
10:30:29amiconnjhMikeS: Nice findings about the PP clock skipping :) But maybe we shouldn't put everything we find out into pp5020.h. It might become really long
10:30:34amiconnWhat do you think?
10:30:56jhMikeSamiconn: Where should we document things?
10:30:59*amiconn could add his findings about the clock setup either there, or to a wiki page
10:31:00obolinuxstb: 7341: usb_detect() in apps/main.c:425 always returns false
10:31:43*jhMikeS guesses if a wiki page gets setup, he'll just move the descriptions to that.
10:31:45pondlifejhMikeS: Do you know why voice causes constant boostage?
10:31:46pixelmaJdGordon: if you choose the latter, I hope the default setting of that would have the same behaviour as it is now - and it doesn't break something else *remembers that horizontal scrolling thing*. (I'm also not sure I like that patch...)
10:32:08jhMikeSpondlife: yes. waiting on the mutex when boosted never does the check to unboost
10:32:15amiconnjhMikeS: Perhaps we should have 2 wiki pages, one for PP502x, another for PP5002, similar to what ipl has
10:32:30jhMikeSworks for me
10:32:31amiconnJust that we now know quite a bsit more :P
10:32:57pondlifejhMikeS: When did that problem get introduced? I'm sure it worked ok at one point...
10:33:00jhMikeSyeah, they've got about a dozen lines worth :P
10:33:16jhMikeSpondlife: no doubt with automatic boost control
10:33:19JdGordonpixelma: of course
10:33:34amiconnjhMikeS: Imo we could remove the separate 5024.h and redefine the Sansa as having PP5022. What do you think?
10:33:39GodEater_I think I asked this before - but 1) Is it okay if I add fast_readline to the plugin API, and 2) Is it documented anywhere? It takes a lot more parameters than the original version.
10:33:56pondlifejhMikeS: I thought it was working ok after autoboost control was added, but got broken later... I may be very wrong of course.
10:34:35pondlifeSo should it be fixed in the scheduler, or just in playback?
10:34:37jhMikeSamiconn: do the other PP5022s have the same multiple GPIO bits? are all interrupt masks the same or should those be config*.h specific?
10:34:58amiconnI'm 99.5% sure they are the same for all PP502x
10:35:06amiconnjhMikeS: Your GPIO0 definitions in there are a duplicate, btw
10:35:09jhMikeSpondlife: I don't see how it could have unless there was a change later.
10:35:42amiconnIt equals GPIO_MASK in pp5020.h
10:35:58amiconnhi interrupts bit 0
10:36:31 Join XavierGr [0] (
10:36:53linuxstbobo: ?
10:37:05jhMikeStrue, but we need at least GPIO0_MASK and GPIO1_MASK, maybe more. Generic GPIO_MASK mask is sort of outmoded
10:37:08 Quit gandhi ("CGI:IRC")
10:37:41amiconnYeah, then rename GPIO_MASK to GPIO0_MASK and add GPIO1_MASK as well to pp5020.h
10:38:08jhMikeSdoes 5020 have GPIO1_MASK?
10:38:09amiconnThe PP5020 also has all the GPIO ports up to L, so it also has the associated interrupts for sure
10:38:32amiconnI wonder whether there is a way to test it
10:38:44obolinuxstb: the usb boot issue - the while(usb_detect()) in init() is never true, even if the usb is connected
10:39:00linuxstbobo: You mean with that patch?
10:39:04 Join The-Compiler [0] (
10:39:11obolinuxstb: no, without
10:39:27jhMikeSamiconn: the emu thinks it's running on somethings else and does not respond to GPIO1_MASK being set so I think maybe that was introduced in 5022
10:39:46amiconnI would rather think it's just missing from the emu
10:39:58amiconnThe emu is based on knowledge about the PP
10:40:13jhMikeSit's not. the running firmware pays no attention. I've modified the emu to test these things.
10:40:19amiconnjhMikeS: If I configure a GPIO for output, enable the interrupt and then flip the bit, does the interrupt fire?
10:41:19jhMikeSI'm not sure. It may depend on if it's the same pin. I've never tried it.
10:41:21linuxstbobo: So the bug is that the usb detection isn't working, so the special case for early USB mode isn't being run?
10:41:23amiconnI would think it does - an easy way to find out about GPIO1_MASK on PP5020
10:41:58amiconnWell, GPIO_INPUT_VAL reflects what's output on the pins configure for output, so...
10:42:44jhMikeSGPIOC bit 3 is set as output but is also the input for AS3514 IRQ at the same time.
10:43:35obolinuxstb: yes
10:43:38JdGordonis there a way to mute the sim?
10:43:41amiconnAha, then some GPIO pins are probably split like some of the coldfire GPIO pins, and others are combined
10:43:45JdGordoncolume doesnt seem to work
10:43:51amiconnJdGordon: nope
10:44:06amiconnThe sim doesn't implement the audiocodec configuration stuff
10:44:21amiconn-> no volume, treble, bass, mute etc
10:44:29JdGordonwhat about with sdl?
10:44:37amiconnI have no idea...
10:44:55 Quit datachild ("- nbs-irc 2.35 - -")
10:45:57obolinuxstb: I put a sleep in before the detection, incase the usb needed longer to init, but that didn't help...
10:46:38pixelmawhat was that about database on Sansa: does it not work at all with files on the microSD or are there only some functions disabled?
10:47:11pixelmaor rather "features"
10:48:04JdGordondircache is disabled
10:48:08n1spixelma: do you have any comments on the langv2 cleanup? or are you ok with me committing it when the last genlang issue gets worked out?
10:48:12JdGordonbah, misread
10:48:16jhMikeSamiconn: that would have to be the case here too. it's set to 1 by the bootloader and is left that way by rockbox but is definitely reads the IRQ state for AS3514
10:49:24amiconnlinuxstb: Do you think we need to dispatch the button interrupts etc to the cop on PP502x?
10:49:52 Quit webguest52 ("CGI:IRC (Ping timeout)")
10:49:58amiconnI'm looking at the PP502x IRQ dispatcher and want to unify it, with just a few ifdefs within the function
10:49:59JdGordonpixelma: the db works fine with the files on the SD card, I tihnkt he reverse lookup for update only works if dircache is enabled
10:50:06*jhMikeS wonders why that's needed
10:50:22amiconnCurrently there are 3 copies of it, and the ipod versions also check for button interrupts on the COP...
10:50:52jhMikeSpixelma: database reads the SD card
10:51:25pixelman1s: I'd like to try a build with the latest patch first, then report back - in general I'd like to see the patch committed soon :)
10:51:29amiconnjhMikeS: Why what is needed?
10:51:41pixelmajhMikeS, JdGordon: thanks
10:52:04jhMikeSamiconn: to dispatch button interrupts to the COP.
10:52:13amiconnYeah, that's why I'm asking
10:53:06*JdGordon found out how to shut up the sim in the code :)
10:53:09 Join The_Compiler_ [0] (
10:53:19JdGordonbut has "playback" at 2x speed because of it
10:53:31*jhMikeS really plans on keeping it relieved of such responsibilities
10:54:11amiconnjhMikeS: I noticed that we don't do so on H10 and Sansa, and think it makes sense to not do it either on ipods
10:54:12JdGordonSlasheri: ping?
10:54:13 Quit The-Compiler (Nick collision from services.)
10:54:44n1spixelma: good, if you test it as it is now be aware that there is a bug in genlang that makes unused strings have the same id as "Yes" in the translations, which confuses rockbox in some places...
10:54:50 Quit The_Compiler_ (Remote closed the connection)
10:54:57 Join The-Compiler [0] (
10:55:05jhMikeSamiconn: I'm not really sure what the COP would be doing in response to them. anything to do with rolo? even then.
10:55:07linuxstbobo: It still seems to me as if the freezing is the real bug - the failure to detect usb early is just highlighting it.
10:55:36amiconnjhMikeS: The respective interrupts aren't enabled on the cop, hence they won't fire...
10:56:15*amiconn goes ahead and unifies
10:56:17obolinuxstb: I wonder if other (non-ipod) targets would freeze if that usb_detect is disabled when usb is present?
10:56:20The-CompilerWhat does ADC mean?
10:56:22jhMikeSamiconn: I also notice the CPU/COP_CTL also should probabably to reset to 0x00000000 on wake. Bit 31 is not cleared automatically.
10:56:29 Join Entasis [0] (
10:56:58B4gderThe-Compiler: analog-to-digital converter
10:57:00jhMikeSThe OF resets to 0x00000000 on wakeup
10:57:13The-CompilerB4gder: thanks
10:58:09 Join bluebrother [0] (i=IhYNze4V@rockbox/staff/bluebrother)
10:59:02pixelman1s: I'll try with english first
11:00:10pondlifeJdGordon: I'd find volume control in the sim useful if you spot anything...
11:00:36JdGordonI didnt look properly, I manaed to fully disable the audio though as a temp hack
11:01:35linuxstbobo: That would be a good test.
11:01:39n1spixelma: there's also a small issue with an #else that shouldn't have been removed in settings_list.c the one between the different antiskip settings
11:02:02pixelmaJdGordon, jhMikeS: has one of you tried the following scenario on Sansa - initialise the database with card inserted, proceed until it's finished (reboot etc.) - then remove card and hit "Update now"?
11:02:24*jhMikeS still want to know how to determine whether IRQ or FIQ woke a core so that the scheduler can just resleep the core and not check threads with FIQ
11:02:29pixelmaFor some reason that doesn't work on my Ondio (shows ongoing disk activity in the status bar until I stop it, but nothing happens and database info in debug gives me "-1% (xxx entries), where xxx is the "old number" - I already reported that to Slasheri quite a while ago
11:02:56*jhMikeS may have but will try again
11:03:14amiconnNow, let's see whether it compiles...
11:03:26linuxstbjhMikeS: Could the ISR set a flag? Or do you think another way exists?
11:04:10 Join barrywardell [0] (
11:04:26*JdGordon thinks he found how to do volume in sdl
11:05:24jhMikeSlinuxstb: I haven't found one (yet)
11:05:57n1sany opinions on me closing FS #7288 it scrolls ok with a bigger font than default or on a smaller screen, do we really need to make the string long enought to scroll on all screens with any font (think ipod video with a 5pixel font...)
11:06:02jhMikeSpixelma: it seems to get the correct number of entries (1093 vs. 1372). I see "-1%" no matter what though.
11:07:07pondlifeLAME experts: Anyone understand if there is actually a bug in - the "gapless is borked" report..?
11:07:31jhMikeSa couple rounds of that and it won't update now :\
11:08:33XavierGris it known that the bookmark feature doesn't work if you start playback by inserting a folder?
11:09:44jhMikeSit had to go through a commit sequence but did inform me about it
11:12:41 Quit My_Sic (Read error: 113 (No route to host))
11:12:52***Saving seen data "./dancer.seen"
11:13:00pondlifeXavierGr: Bookmarks have quite a number of bugs...not heard that one before though.
11:13:26pondlifejhMikeS: has a "Works for me" fix on it - sleep(1) before waiting for the mutex.
11:13:27pixelmajhMikeS: ok, seems like a general problem with multivolume then (I can't see the full line with entries on my Ondio as the line is too long...
11:14:00obolinuxstb: unfortunately I only have a 5g...
11:14:52amiconnjhMikeS: Not scheduling button ints to the cop is working fine on mini G2. Will test on video as well, then commit
11:14:53jhMikeSpondlife: I thought about that one. It should be reboosted next time voice runs I guess.
11:15:20pondlifeIt seems to work ok on H340. The sleep is in an ok place? I didn't want to break the mutexing!
11:15:46pondlifeOr maybe sleep(1) rather than yield() in the loop?
11:16:20jhMikeSpondlife: I forgot, the swap will unboost during clip playback with that.
11:16:22pixelmaXavierGr: there are a few restrictions with bookmarks and dynamic playlists currently. I remember it being discussed a while ago ( at 10:53 o'clock and following)
11:17:01pondlifejhMikeS: Is that a problem? It seems to work ok - no skipping or under-run.
11:17:28pondlifeBetter than the reduced battery life that SVN gives, I'd think.
11:17:48jhMikeSvoice should stay boosted until the clip is finished really.
11:18:18pondlifeHmm, so sleep somewhere else...
11:18:37amiconnpondlife: Keep in mind that we don't want to boost/unboost in too quick succession
11:18:53jhMikeSI'm not sure where.
11:19:31pondlifevoice_pcmbuf_insert_callback() when it's out of data?
11:20:05pondlifeOr is that in an ISR?
11:20:53pondlifeHmm, the queue_wait_w_tmo ()in voice_request_buffer_callback() ought to handle this anyway, no?
11:21:12jhMikeSyou can't really tell if it is there
11:21:13XavierGrpondlife: about 7334, follow playlist does work for me even on database mode
11:21:32pondlifeXavierGr: How is that possible?
11:21:36jhMikeSpondlife: no, not when voice is swapped out
11:21:45linuxstbobo: OK, I can try that (I have a gigabeat and h140)
11:22:02XavierGrpondlife: ah wait maybe I misunderstood let me test something
11:22:08pixelmaXavierGr: when music is stopped?
11:22:43pondlifeXavierGr: It could go to "All tracks", but otherwise there's no way to know which virtual folder you started with - could be one of many.
11:22:58XavierGrfalse alarm I was tricked by the fact that the database remembers where you were last time
11:23:28pondlifeI think that "Follow Playlist" should mean that STOP or SELECT go to the file browser, even if you were last in the database...
11:23:51 Join PaulJam [0] (
11:24:21pondlifeThat would seem consistent - if you want to use the database only (e.g. iTunes) then don't use Follow Playlist.
11:25:36jhMikeSpondlife: voice_is_playing should be false after the swap_codec call in voice_pcmbuf_insert_callback if voice finished, right?
11:25:56pondlifeYes. But playing remains true (of course)...
11:27:43jhMikeSpondlife: right after voice_is_playing = false in voice_on_voice_stop probably
11:27:53pondlifejhMikeS: Alternative fix - in voice_request_buffer_callback(), change the first queue_wait_w_tmo(...,0) to be (...,1).
11:29:16*pondlife would like the status bar to have a "CPU boosted" marker.
11:29:17jhMikeSno, that'll unboost on each frame
11:29:23amiconnB4gder: It's not correct that rockbox wasn't ported to telechips hardware yet. Just that the port had died and was removed again...
11:29:28 Join webguest52 [0] (i=c0231115@gateway/web/cgi-irc/
11:29:56jhMikeSif (!voice_thread_start) sleep(1)
11:30:11pixelmajhMikeS: in addition to the "update now" - I chose a tinier font and can now read the complete "progress" line. It gives me only one number of entries which I even cannot recognise when looking at the numbers the "stats" plugin gives me
11:30:16pondlifeor if (!voice_is_playing) sleep(1) ?
11:31:13B4gderamiconn: true, but that telechip one is bound to be quite different to those newer ones anyway
11:31:14jhMikeSno, because voice_is_playing will be true most likely
11:31:31jhMikeSthat's where it's set to false in the first place
11:32:12 Quit webguest52 (Client Quit)
11:32:17amiconnB4gder: Yes it is, but the question was whether there was an attempt to port to telechips hardware, not what telechips hardware that was...
11:32:34 Join helloworld1 [0] (n=pocketir@
11:32:50B4gderyeah, I forgot that the gmini one as a telechip
11:32:51jhMikeSthat should make sure it doesn't unboost if more clips are requested
11:32:55 Join webguest52 [0] (i=c0231115@gateway/web/cgi-irc/
11:33:56pondlifejhMikeS: Feel free to commit something :)
11:34:12pondlifeI've got to do work now :(
11:36:43jhMikeSok...I'll test some stuff out then
11:37:28 Quit helloworld1 (Remote closed the connection)
11:39:30pondlifeThanks muchly... I'm sure you have to do work sometimes too..!
11:39:48jhMikeSindeed, but my hours are odd :)
11:41:27 Quit darkless ("Leaving")
11:41:38 Join darkless [0] (
11:45:57B4gderthe char that killed convbdf was a whopping 52 x 39 ;-)
11:47:39 Quit Rob222241 ()
11:48:27peturXavierGr: I'm not convinced that 6614 itself is fixed, it is just that the way he arrived at the bug got fixed. I don't know if we allow (or should allow) min and max dB settings to be the same.
11:48:50PaulJamlittle question about buttonactions in plugins: could it be that there is currently no action that uses the REC button on H300?
11:50:23XavierGrpetur: but I can no longer reproduce it? anyway if you want me I will reopen it
11:50:48peturXavierGr: ah, so you tried setting min and max to 0dB?
11:51:25peturanyway, it's not an important use case ;)
11:54:22XavierGrwell at 0db maximum/minimum there are no longer any peak meters to be shown, but I think that this is a very logic reaction
11:56:02XavierGrwow the peak release menu shows to my gigabeat quite some garbage after the number entries!
11:56:09 Quit linuxstb ("Leaving")
11:57:05XavierGrok the peak release menu is majorly bugged
12:00:50 Part maffe
12:01:03 Join maffe [0] (
12:01:05JdGordonPaulJam: yes
12:04:53JdGordonPaulJam: I got volume working, :)
12:05:32PaulJamJdGordon: what?
12:05:47JdGordoncrap, that was for pondlife, the first one was for yout hough
12:06:06JdGordonhow do i go from a number between -40 and 6, to a number between 0 and 128?
12:06:42peturax+b ?
12:06:54 Join Rob2222 [0] (
12:07:05*JdGordon knew grade 6 maths would come back to haunt me :p
12:11:47JdGordonthanks :)
12:12:01pondlifeJdGordon: thanks :)
12:17:50JdGordonwell... change all that to use VOLUME_MIN and MOLUME_MAX and its perfect :)
12:18:39jhMikeS128*(x - VOLUME_MIN)/(VOLUME_MAX-VOLUME_MIN)
12:22:09 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
12:22:29n1sJdGordon: I found a way to fix FS #7311 it is really weird to me though...
12:23:21n1sboth the clock and vu_meter plugins stored their settings in structs called "settings" change the name of the struct and voila, no crash...
12:23:48JdGordonname conflict?
12:24:09maddlermorning all...
12:25:00n1sJdGordon: well, the crash was at line 390 in settings.c so i thought it seemed probable... and it works...
12:25:26n1sbut why would it only cause problems on 64 bit?
12:29:17n1sJdGordon: so should I commit it? or do you want to look into it and make a proper fix?
12:29:43JdGordonI'm 32bits short, so may as well commit it
12:30:50*JdGordon back in a while
12:32:33 Join dan3333 [0] (i=cbd63214@gateway/web/cgi-irc/
12:33:45 Quit dan3333 (Client Quit)
12:38:07pondlifejhMikeS: Nice, I'll test and close
12:41:21 Join daniel1234 [0] (
12:42:47daniel1234hi, I want to start coding on rockbox. I have made the total disk size speak but I want to try and make some splash messages speak. How should I do this? And how do the talk functions work?
12:43:37jhMikeSI think this auto boost should be something that has to be maintained or else it just expires
12:43:51amiconnI managed to run the PP5002 at 32kHz :)
12:44:36amiconnMany rockbox functions would block when they are called at that clock frequency, because the microsecond timer doesn't run
12:44:37jhMikeS:) /me hopes the codecs become more optimized :P
12:45:04amiconnThis is for fake shutdown of course
12:45:15daniel1234Also what do I need to know in c to be able to start coding on the voice stuff
12:45:43amiconnI tested that I'm really running at 32kHz by timing a known delay loop with my watch
12:45:56amiconn32768 loop cycles took ~4 seconds
12:46:47amiconnMake that 16384, but you get the idea...
12:47:32jhMikeSwhat's "fake shutdown"? the thing apple firmware does?
12:47:40pondlife"deep sleep"?
12:48:02amiconnThe ipod 1st/2nd gen can't be shut down physically
12:48:09amiconnThe PP5002 is powered all the time
12:49:42amiconnApple just sets it to 32kHz and disables the lcd
12:49:55amiconnWe will need to do the same
12:51:33 Quit webguest52 ("CGI:IRC (Ping timeout)")
12:52:19 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:03:35 Quit PaulJam (".")
13:12:56***Saving seen data "./dancer.seen"
13:17:43daniel1234How do I register xvoice.dll 
13:18:03 Join Nick_Brackley [0] (i=cb57494b@gateway/web/cgi-irc/
13:18:24n1sdaniel1234: google for regsvr32
13:19:57daniel1234ok and I am using vista is that ok\?
13:20:21n1sdaniel1234: dunno, never touched vista
13:20:28daniel1234do I just google for "regsvr32 "
13:21:09n1sdaniel1234: yes
13:21:27n1sit will giv eyou a microsoft page that tells you how to use it
13:25:01 Join webguest36 [0] (i=c0231115@gateway/web/cgi-irc/
13:25:22pixelmajhMikeS: "update now" also struggles when I insert a different MMC...
13:27:45 Join Thundercloud [0] (
13:29:14 Quit webguest36 (Client Quit)
13:30:10 Join webguest36 [0] (i=c0231115@gateway/web/cgi-irc/
13:30:24pixelmahmm... but it did create new database entries. Just because it can't check for deleted files (missing dircache) it just added the new entries to the old one. Seems like this is expected behaviour then...
13:36:52 Quit daniel1234 ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
13:37:10barrywardelljhMikeS: I think your commits 14035-14037 somehow broke the h10 build. I'm getting a data abort on startup now
13:37:11jhMikeSpixelma: I suppose "Initialize Now" is what would do a full rebuild?
13:37:44barrywardellat 00047D44
13:37:53jhMikeSbarrywardell: will check. which were those?
13:38:01pixelmajhMikeS: yes (afaik)
13:38:07pixelmaand what I do
13:38:49barrywardelljhMikeS: Scroll on main and remote with a single thread. Change the way system messages are defined before running out is an issue (which requires a full update of rockbox on the player).
13:39:41 Quit webguest36 ("CGI:IRC (Ping timeout)")
13:40:06jhMikeSbarrywardell: all the time?
13:40:28barrywardelljhMikeS: and here's my
13:40:50barrywardellyes, every time I start up, before I see root menu
13:41:26jhMikeSso the problem starts on exactly that revision?
13:42:01barrywardellyes, 14034 works fine. I don't even see the rockbox logo before getting the data abort
13:42:27 Quit safetydan ("Ex-Chat")
13:42:48 Join freqmod [0] (
13:48:06JdGordonamiconn: shouldnt the entire system be put into some infintie loop while its sleeping? instead of having to find and fix things which obviously wont work at such slow speeds?
13:48:48amiconnOf course
13:49:13amiconnActually, the COP should be put to sleep entirely, and the CPU too with just leaving the button interrupts enabled
13:50:16 Quit daurn (Connection timed out)
13:50:19pondlifeJdGordon: red...
13:50:35pondlifeOr did you already know?
13:50:39JdGordonyeah, i know
13:50:42JdGordoncommiting in a sec
13:50:46*JdGordon forgot */
13:51:10B4gder1678 fresh points for JdGordon
13:51:24JdGordonthats not even close to a high score for me :D
13:51:34pondlifeGet 'em while they're hot, they're lovely...
13:52:10pondlifeAnd more commits will helps sort out the build table.
13:52:59jhMikeSamiconn: does your H10 bomb out at startup now?
13:53:05pondlifeJdGordon: I knew there was a reason not to comment code... :)
13:53:17 Join Siltaar [0] (
13:53:19JdGordonhaha yeah
13:53:31pondlifeanyway, thanks for the volume control
13:53:42amiconnjhMikeS: Nope, works fine
13:54:01amiconnI am not running the latest, but r14056
13:54:44amiconnbarrywardell: Did you perform a full rebuild, or optionally try an official build?
13:55:06barrywardellI tried a full rebuild, trying an official build now...
13:55:11jhMikeSthe system message values did change
13:55:41JdGordonwasnt the stack increased so mpegplayer doesnt crash? or am i dreaming it?
13:56:08*jhMikeS noticed a funky thing that should be fixed up: lcd_enabled always returning true on everything but gigabeat :\
13:56:17pondlifeJdGordon: No, but a fix was applied.
13:56:36JdGordon seems not...
13:56:36barrywardellhmmm. official build works
13:57:09barrywardelljhMikeS: sorry for the false alarm. must be something wrong with my build env
13:57:09pondlifeJdGordon: Is that a stack overflow?
13:57:15jhMikeSscore one for me! \o/ The H10 didn't win this round :)
13:57:42JdGordonpondlife: no, but i thought the crashes wernt always stkovs?
13:57:58pondlifeTrue, but those might not be.
13:58:24pondlifeGodEater: DevConPub3 already? ;)
13:58:48pondlifeMaybe in September/October sometime...
13:59:01barrywardellafter deleting my build dir and doing a fresh compile, everything works again
14:00:08jhMikeSwhew. PP5020 having trouble makes /me panicf
14:00:36 Join webguest36 [0] (i=c0231115@gateway/web/cgi-irc/
14:04:34JdGordonshould 7473 be committed assuming of course it doesnt cause crashes?
14:05:09jhMikeSamiconn: what file is the lcd driver for 1g/2g?
14:05:25*JdGordon wonders if people prefer the full tracker url or just the task number?
14:05:33pondlifeFull URL here
14:05:39amiconnfirmware/target/arm/ipod/lcd-gray.c, why?
14:05:43pondlife7473 ism a bug, not a patch.
14:06:01jhMikeSamiconn: just a small addition for scroller compatibility for HAVE_LCD_ENABLE
14:06:33amiconnsvn is running fine on my 2nd gen...
14:06:52*petur managed to stop clicking the #abcd links in his irc client
14:07:00jhMikeSsomeone added lcd_enabled() to gigabeat and just #defined it as lcd_enable() true for all others
14:07:32amiconnThe lcd is always enabled on the ipods...
14:07:53JdGordonpondlife: no its a patch....
14:07:55amiconnWe don't know how to disable it yet, and for grayscale it doesn't make sense
14:08:06pondlifeJdGordon: ?
14:08:09jhMikeSwhy is #define HAVE_LCD_ENABLE in the config for 1g,2g,3g then?
14:08:16amiconn(except for sleep but that's an entirely different thing)
14:08:29 Join rogelio [0] (n=rogelio@
14:08:31*JdGordon dyslexik tonight...
14:08:58amiconnI dunno; didn't put it there
14:09:15linuxstbbarrywardell: I had a similar problem with my ipod Color recently - starting from a clean build directory was necessary for me as well.
14:09:15amiconnThe 1g/2g config started as a copy of the 3g
14:09:26 Quit webguest36 ("CGI:IRC (EOF)")
14:09:32 Join Febs [0] (
14:09:34 Part rogelio ("Kopete 0.12.4 :")
14:09:46jhMikeSwhich driver is 3g?
14:09:50pondlifeJdGordon: I say it should be ok...but check by skipping backwards and very quickly deleting the file?
14:10:10jhMikeSshould be the same?
14:10:20amiconnyes. 4g too
14:10:25barrywardelllinuxstb: ah, ok. that must have been the problem in that case
14:10:30amiconn...and the minis
14:10:59jhMikeSsame, but I see no lcd_enable implementation there. probably should go then.
14:15:32*JdGordon tihnks he accidently deleted 2 flac cds :'(
14:18:13pixelman1s: I'm running a langv2 patched version on my Ondio (with backlight enabled) - Got a "Not done yet!" in the playback settings where the "Anti-skip buffer" is now... since I guess it's because that one would be removed for Ondios now (IIRC) it is only a bit confusing. Will try to use it a bit longer "normally" and browse through the settings...
14:20:07JdGordonpondlife: hmm, doesnt seem to cause problems seeking past the end of the buffered part of the file
14:20:50linuxstbamiconn: Do you know how to set the clock to 32KHz on the PP502x as well?
14:21:24pondlifeJdGordon: The sim volume control works well, but the H300 sim allows up to +6dB, unlike the target...
14:21:28amiconnlinuxstb: yes
14:21:45JdGordonpondlife: all the sims have the same limit
14:21:51pondlifeAh, ok.
14:22:04 Join joshin [0] (n=joshin@VDSL-130-13-8-23.PHNX.QWEST.NET)
14:22:07pondlifeI'd have expected them to be more like their targets.
14:22:10linuxstbamiconn: Are you interested in implementing a sleep option there as well?
14:22:32n1spixelma: try adding a #else between the two different antiskip options in settings_list.c it was accidentally removed
14:22:49amiconnI won't implement a sleep option on PP5002, just fake shutdown
14:23:09linuxstbWhat's the difference? I'm using the term "sleep" to mean what Apple's firmware does...
14:23:16pixelman1s: ok, will do
14:23:18JdGordonany last objections to (yay for autoreplace script)
14:23:27amiconn...i.e. disable everything in rockbox, then put the PP5002 to sleep. A button press will cause a reboot
14:23:46barrywardellamiconn: is it just a matter of setting bits 0-2 of CLOCK_SOURCE to 0 (=32kHz)?
14:23:48n1spixelma: ok, please post any other issues either here or in the tracker, gtg now
14:24:22peturJdGordon: killing two tracker entries at once ;)
14:24:23pixelmasure - would have done it anyways if I didn't see you around here
14:24:37amiconnbarrywardell: Probably that plus some delay in order to let the 32kHz stabilize before switching to it
14:24:48amiconnCheck what I did on PP5002
14:24:58 Join webguest36 [0] (i=c0231115@gateway/web/cgi-irc/
14:25:48amiconnWe also need to find out how to disable the various hardware components, otherwise the ipod will still drain more power than necessary in sleep mode
14:28:19linuxstbJdGordon: Did you give up trying to pass a bootloader version to Rockbox?
14:28:51JdGordonyes and no, I went on to differnt bootloader fiddling and forgot about that
14:29:06JdGordonbut I couldnt get that working which annoyed me
14:29:26bluebrotherdo we really need another item in the context menu? *sic*
14:29:46linuxstbI'm thinking about implementing a "Start original firmware" option (for the System menu) on ipods, and for that I'll need to know if the bootloader supports it...
14:30:49linuxstbJdGordon: Also, about your last commit - did you test deleting a large file - i.e. one that hasn't been fully buffered?
14:31:00JdGordon130mb one big enough?
14:31:12linuxstbYes ;)
14:31:17*linuxstb is being blinded by the sun
14:31:42linuxstbI just recall lots of people saying that deleting the currently playing file didn't work reliably. (I've never tried myself though).
14:31:50JdGordonbluebrother: let me commit customizablemenus and you can have as few as you want
14:32:04bluebrothernoooo ...
14:32:18JdGordonwell, the patch is tiny, so if it does start causing problems it can be removed painlessly
14:32:30JdGordonbut I did test and came up fine, so thought what the heck :)
14:32:48JdGordonat a cost of a whopping 4 bytes :p
14:32:51pixelmaJdGordon: you could also help documenting those new "features" in the manual...
14:33:07linuxstbJdGordon: The other objection to that patch was that it could confuse users who might think it just means deleting the track from the playlist...
14:33:46JdGordonpixelma: you really want me to fiddle with the manual? imagine the chaos with speeling and macros i can do with TeX? :D
14:34:12JdGordonlinuxstb: ah, well thats a good point, meh
14:34:17bluebrotherJdGordon: you really sure you can cope with TeX' macros? :P
14:34:22linuxstbJdGordon: If you don't want to touch the manual, it could be useful to add flyspray tasks to keep track of the missing features.
14:35:13JdGordonbluebrother: I did a tech doc subject beginig of last year which taught us LaTeX and it scarred me for life
14:35:26bluebrotheryou could also add desctiptions of the new features, along with noting the devices it applies to.
14:35:45bluebrotherJdGordon: ever looked at MusixTeX? *That* is scary.
14:36:09barrywardellamiconn: I see. iiuc it would just need to have the numbers for CLOCK_SOURCE changed (the source frequencies map differently on PP502x?)
14:37:00amiconnI should really start that wiki page and publish my findings (based on MrH's work)
14:37:37pixelmaJdGordon: about the spelling... the wiki pages (s) you create ofte read quite nicelyl - you're just typing too fast in IRC
14:37:45amiconnThe clock setup on PP502x is basically the same as on PP5002 but with the bit shuffled around and some extensions
14:38:01pixelmasorry about the spelling :P
14:39:12barrywardellcool, it should be very easy to add CPUFREQ_SLEEP
14:43:11amiconnE.g. PP5002 has 4 clock mode and PP502x has 8
14:44:37barrywardellah, the code makes more sense now. I was expecting 3 bits for each source, but for PP5002, there are only two
14:46:54bluebrotherJdGordon: take a look at this ... ;-)
14:47:42amiconnbarrywardell: Btw, is your H10 stable with the new clock setup function? My H10 6GB is :)
14:48:01bluebrotherthat's scary (and I did write it)
14:48:10barrywardellyes, it seems to be very stable. I haven't done very extensive testing, but all seems good so far :)
14:48:19linuxstbbarrywardell: You have the 20GB/
14:48:51barrywardellthey're both very similar. the biggest difference is the lcd
14:48:52amiconnThe 2nd gen is also 100% stable, so I'd expect the 3rd gen to be stable as well :)
14:49:29*linuxstb should start using his ipod color more often, especially as he claimed to be in the MAINTAINERS file...
14:54:13 Quit XavierGr (Nick collision from services.)
14:54:16 Join XavierGr [0] (
14:58:29 Join Jens [0] (i=Jens@pdpc/supporter/active/Jens)
14:59:56JdGordonjhMikeS: your recent scrolling hacking means can be out-of-dated right?
15:00:04 Quit webguest36 ("CGI:IRC")
15:01:08jhMikeSJdGordon: lemme look
15:01:40JdGordonbtw, red if you didnt see
15:02:11jhMikeSI saw. fixing it.
15:02:30jhMikeSIt's partially outdated. I didn't move any puts functions.
15:04:49amiconnbarrywardell: When comparing the H10 and iPod color lcd drivers, you'll see that they both use the same lcd bridge in the PP
15:05:10amiconnSo it should be possible to port the multiple-write optimisation from iPod color to H10
15:05:23barrywardellyes, I noticed that. That's how I got the H10 LCD working :)
15:05:24amiconnShould be good for another ~30% lcd speedup....
15:05:42barrywardellThat, and the fact that the LCD is identical to the X5
15:05:52 Quit idnar (Nick collision from services.)
15:05:54barrywardellI'll have a look at that optimisation
15:05:54 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
15:05:57amiconnYes, the H10 large
15:05:59jhMikeSJdGordon: actually, all bitmap displays could share the same scrolling code on both remote and main. I didn't go that far because I thought it too much in one go for that.
15:06:18amiconnThe H10 small has TL1772, but the bridge is still the same
15:07:01JdGordonjhMikeS: righto, ill have a quick look and see if im awake enough to do it
15:07:02amiconnI didn't find the TL1772 datasheet (seems to be secret), but I found a datasheet of an LCD module that uses TL1772
15:07:13barrywardellhow does DEV_TIMING1 work? What should it be set to for 32kHz?
15:07:21amiconnIt lists all commands plus the sleep/unsleep sequences etc
15:07:57barrywardellgreat, sleep still needs to be enabled for it. It's working on H10 20GB already
15:08:08amiconnI don't know. I'd try to either set it to minimum (0x00000101) or leave it alone...
15:08:32amiconnbarrywardell: I know... I'm busy with PP5002 atm, otherwise I'd add that to the H10 driver...
15:09:12barrywardellok, i'll leave it alone for the first test
15:10:03amiconnDon't forget to disable interrupts for your test
15:11:43barrywardellwill do. how did you time the test?
15:12:41jhMikeSJdGordon: I did implement it that way at first but then deleted all that. :P Get's kinda ugly with different color depths if you can't use the screens API down there.
15:13:00***Saving seen data "./dancer.seen"
15:14:00JdGordonall these lcd_puts* functions dont really belong in scroll_engine.c though, I'm dumping them in lcd-bitmap.c and seeing how bad it is with #ifdef HAVE_LCD_COLOR everywhere
15:15:41 Quit Nick_Brackley ("CGI:IRC (Ping timeout)")
15:16:53 Join XavierGr_ [0] (
15:16:53 Quit XavierGr (Nick collision from services.)
15:19:24JdGordondoes the player have a lcd remote?
15:21:54jhMikeSI think only iRivers and iAudios have them...the coldfire ones
15:25:24*jhMikeS does a double-take a sees two columns for iPod 1G/2G
15:26:25pondlifeOne more commit...
15:26:52 Join MacFra [0] (i=570fbad5@gateway/web/cgi-irc/
15:27:07*pondlife invites you to play Build Table Tetris
15:27:13MacFrahi guys!
15:27:46jhMikeSthe build table is getting wider by the minute lately
15:28:01pondlifeMight need a bigger monitor...
15:28:09JdGordonneeded on months ago already
15:28:09MacFracan you please tell me what build is right for ipod 5.5 80 gb??
15:28:11peturor a smaller font
15:28:33pondlifeMacFra: iPod Video 64MB
15:28:36JdGordonjhMikeS: added a quick patch to merge the lcd_puts* functions if you want to look at it, goig to bed so not finishing it tonight
15:28:37B4gderbuild-table tic-tac-toe, you get points if you get three reds in a diagonal!
15:28:48linuxstbMacFra: The 64MB one.
15:28:54pondlifeConnect 4 (reds)
15:29:18B4gderplain horizonal or vertical is too easy ,-)
15:29:38jhMikeSJdGordon: oki
15:30:13 Part n1s
15:30:24*jhMikeS could make it all red :D
15:30:26MacFraok tnx
15:30:38barrywardellamiconn: when does rockbox decide between CPUFREQ_DEFAULT and CPUFREQ_NORMAL?
15:30:40*pondlife has made it all red...
15:31:01jhMikeSpondlife: every cell in every row?
15:31:06pondlifeAh no!
15:31:17pondlifeThat wouldn't be hard though
15:31:29pondlifeYou've sunk my battleship!
15:32:54jhMikeSyou pick a row and column and I'll guess by committing exactly one red build at a time
15:33:11JdGordonthats harder than it soounds though
15:33:13 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
15:33:16JdGordoncommiting one red cell at a time...
15:33:43peturnot if you understand the art of config_xxx.h
15:33:45linuxstb#if defined(IPOD_3G) && defined(BOOTLOADER) ....
15:34:01JdGordon... unless you cheat like above!
15:34:03jhMikeShence the challenge. If I commit more than one, you have to mark those cells with warnings only.
15:34:39peturmaybe as an extra Bagder can rename builds to give it some extra effects
15:34:44JdGordonthis is rockbox' contribution to global warming :) keeping those servers busy
15:35:39jhMikeSit'll need a nuke plant to run it soon and no CO2 emissions and loads more fun with radiation
15:35:54 Quit MacFra ("CGI:IRC (Ping timeout)")
15:35:56amiconnbarrywardell: CPUFREQ_DEFAULT isn't used at all on PP except in the cpu frequency debug screen
15:36:11amiconnOn coldfire it's used in the radio screen to save some more power
15:36:29amiconnI timed the test with an asm delay loop running 16384 cycles
15:36:52amiconnTook ~4 seconds on PP5002
15:37:03amiconnOn PP502x it should be a bit faster, 2 or 3 seconds
15:37:35pondlifejhMikeS: Nuclear? So a good build won't just be green, it'll be luminous.
15:37:42 Join webguest19 [0] (i=570fbad5@gateway/web/cgi-irc/
15:38:06jhMikeSa nice warm glow
15:38:27pondlifeDo we now believe that all PortalPlayer targets are stable? Or is it too early to say?
15:38:32webguest19to install rockbox
15:38:37webguest19sorry again
15:38:45webguest19do i need to update all my songs anywhere else?
15:38:46barrywardellamiconn: cool, thanks
15:38:47linuxstbpondlife: I think some Nanos and Videos are now unstable...
15:39:07*pondlife fails to do the PP stability dance... or clear up loads of FS entries :(
15:39:38JdGordongnite folks
15:39:43 Quit JdGordon ("Konversation terminated!")
15:39:58amiconnbarrywardell: I.e. print something to the lcd, lcd_update(), disable interrupts, switch to CPUFREQ_SLEEP, run delay loop, restore clock, restore interrupt setting, print something and lcd_update() again
15:40:41jhMikeScould use the cycle skip method too now :)
15:40:47barrywardellamiconn: exactly what I plan to do :)
15:41:22amiconnjhMikeS: Is the cycle skip applied to every instruction, or just once after setting the register?
15:42:02jhMikeSamiconn: just the one following setting the reg
15:42:21amiconnaha, hmm
15:42:33*amiconn can't imagine what this is good for then
15:43:28pondlifelinuxstb: You think some Nanos and Video are now unstable...? Any more clues?
15:43:39amiconnHow does that help with mutexing?
15:43:45 Quit webguest19 ("CGI:IRC (Ping timeout)")
15:44:36jhMikeSyou stall the other for a certain number of cycles if you're doing something not atomic so that it can't execute the instuctions that the unstalled cpu could in that time frame. it seems to actually be used like that.
15:45:25linuxstbpondlife: See the forums and IRC logs....
15:46:21pondlifeHmm, I wasn't sure if the forum reports were still relevant or not. Maybe they are not...
15:46:44*linuxstb looks to amiconn to describe known PP issues to pondlife
15:47:13pondlifeIt's ok, I don't need the full detail... just was hoping to blitz some more "music playback" bugs.
15:47:22pondlife...that are more "PP" bugs.
15:48:41pondlifeI was hopeful that having seen amiconn mention 100% stability on 2G/3G, then he may have found out some useful stuff for the later models too.
15:51:37linuxstbIIUC, PP5020 is now stable, but there are IDE timing issues on PP5022
15:54:14*barrywardell has a working CPUFREQ_SLEEP on his H10 :)
15:54:38*amiconn just managed to crash his 2nd gen
15:54:45pondlife99.9% ?
15:54:59amiconnNo, I did something I shouldn't have done
15:55:18 Quit Febs (Read error: 110 (Connection timed out))
15:55:18amiconnI entered the CPU_SLEEP test while music playing, but my test doesn't disable fiq, only irq...
15:56:06linuxstbbarrywardell: Thinking about a sleep mode?
15:56:47amiconnI wanted to enter 'View battery' but accidentally entered 'View HW info' which does that test...
15:56:54barrywardelljust keeping up with amiconn for now :)
15:57:24*linuxstb hands barrywardell some fast skates
15:58:14*pondlife hands over a jet pack too
15:58:17linuxstbbarrywardell: BTW, have you tested Rolo recently on your H10? Has it been adjusted to .mi4 files yet?
15:58:57*GodEater_ finishes working in his changes to make "shortcuts" use read_line() and allow users to mark files as well as directories.
15:59:04barrywardellI think I adjusted it for mi4 files a while ago IIRC
15:59:25GodEater_now - any clues as to why my diff thinks I've changed masses of stuff in onplay.c when I haven't ?
15:59:28*barrywardell accepts pondlife's jetpack but doesn't think it will be enough
15:59:56*pondlife fears barrywardell will crash, not just his H10.
16:00:05midkayGodEater_: whitespace changes perhaps?
16:00:39GodEater_midkay: I didn't make any
16:01:01GodEater_the diff seems to be pulling out blank lines, and then putting them back in again - it's crazy
16:01:02linuxstbDid your editor? e.g. some might do things like strip trailing spaces
16:01:07midkayare you sure your editor didn't?
16:01:13GodEater_not to my knowledge, but I'll check
16:01:22midkayyeah, right. some might remove useless spaces or convert tabs to spaces..
16:01:57*pondlife is loving the random album feature.
16:01:57linuxstbbarrywardell: Was rolo loading the original firmware (on either Sansa or H10) ?
16:02:04bluebrotheryou might try diff -b to ignore whitespace changes
16:02:40*bluebrother wonders how to pass options to svn's diff
16:02:46GodEater_well we're supposed to not have tabs right ?
16:03:02linuxstbGodEater_: My guess would be trailing spaces...
16:03:11barrywardelllinuxstb: I can't remember, but I think it half worked (it started loading, but not fully)
16:03:26pondlife"Blank" lines which contain 4 spaces are quite common.
16:03:36GodEater_ah - passing -b does indeed cut the size of the patch down massively
16:03:36linuxstbbarrywardell: OK. I want to try and get it working on ipods, so was curious about its state on H10/Sansa.
16:03:41pondlifeIn SVN, I mean.
16:03:45midkayor line endings? CR/LF stuff or whatever?
16:04:10barrywardelllinuxstb: I'll try it again in a few minutes and let you know
16:04:11GodEater_bluebrother: I used git diff -b master..HEAD to get my diff - seems to pass the -b okay
16:04:18*amiconn does to implement dump_ide_registers()
16:04:25GodEater_definitely wasn't line endings - I checked that
16:05:06bluebrotherwell, git diff can also output in color −− quite a nice advantage compared to svn
16:06:40GodEater_not much use in a patch though ;)
16:06:55 Join Febs [0] (
16:07:01linuxstbYou use "−−diff-cmd xxx" to use an external diff utility with SVN.
16:09:02 Join XavierGr [0] (
16:09:38 Join Soul-Slayer [0] (n=jonno@
16:12:18 Join topdog101 [0] (i=c4c028c7@gateway/web/cgi-irc/
16:13:37pixelmalinuxstb: did you read the update of the one reporting instabilities on his Ipod Video in today's log (somewhen in the morning)?
16:15:02GodEater_pixelma: was that the chap who may have been short changed by Best Buy ?
16:16:14linuxstbHe was the only person reporting problems on the Video?
16:16:33pixelmaGodEater: yes
16:16:43 Quit topdog101 (Client Quit)
16:17:00pixelmalinuxstb: I'm not sure - but that's the only one I can think of ATM
16:17:13 Join desowin [0] (
16:18:54 Quit XavierGr_ (Read error: 110 (Connection timed out))
16:23:26 Join low_light [0] (i=c730180a@gateway/web/cgi-irc/
16:24:22 Join webguest22 [0] (i=4452cf40@gateway/web/cgi-irc/
16:24:48webguest22Does anybody know why I can't get Flac format songs onto my Ipod?
16:25:03Soul-SlayerCopy and paste them onto the drive in disk mode
16:25:08Soul-SlayerThere's no reason they shouldn't copy.
16:25:16bluebrotherbecause Itunes doesn't support flac?
16:25:19B4gderwebguest22: sure, don't tell any details we'll soon have it figured out!
16:25:33*bluebrother leaves for his crystal ball
16:25:49webguest22I use foobar, and just rockboxed my 80 gig ipod
16:26:00webguest22and when I try to send it to my ipod, it says flac not supported
16:26:12B4gderwebguest22: just drag and drop the files
16:26:17bluebrotherwell, Ipods usually don't support flac ...
16:26:20B4gderno need to complicate matters
16:26:36bluebrotherso (unsurprisingly) any Itunes sync plugin won't sync flac files.
16:26:40webguest22ok, anywhere in the disc or the main folder?
16:26:40B4gderand then you go whine to the foobar guys :-)
16:26:48B4gderwebguest22: anywhere you like
16:26:56Soul-SlayerBut organisation always helps
16:27:13Soul-SlayerHaving a root folder flooded with .flac files could get messy :p
16:27:14B4gderright, at least when organisation means not everything in a single folder ;-)
16:27:27barrywardelllinuxstb: rolo-ing rockbox works, rolo-ing the of doesn't (on my H10)
16:27:30bluebrotherI usually keep all my music below a folder /Music. Others may have different structures.
16:27:36webguest22could I just copy paste my harddrive onto it?
16:27:42webguest22witha ll my music
16:27:48bluebrotherthis also makes it easy creating a playlist from all files ;-)
16:28:08webguest22really I just want them to be sorted by artists then album and I'm good, so thanks
16:28:09Soul-SlayerYou can just copy and paste your music folder onto your ipod and it will be recognised.
16:28:10bluebrotherthe Ipod is simply a removable disk when running Rockbox.
16:28:17*amiconn recommends bluebrother to try this on a FAT16 volume ;)
16:28:50linuxstbbarrywardell: You may want to do a test - dump the contents of IRAM to disk from the bootloader. The Apple bootloader seems to pass various bits of info to the main firmware via a struct at the end of IRAM.
16:29:07Soul-Slayerpondlife: Are we concerned that crossfade doesn't trigger now if you pause -> skip song -> play? There are no adverse effects apart from it simply not crossfading, however.
16:29:16*bluebrother wonders what amiconn wants to point out
16:29:46 Quit webguest22 ("CGI:IRC (EOF)")
16:30:34amiconnbluebrother: The FAT16 root dir is fixed size. 512 entries, where a long filename takes multiple entries (up to 6 iirc)
16:31:13bluebrotheraaah. I know that fact but I was wondering what I should try −− I don't put all my music in the root folder ;-)
16:32:42oboB4gder: make mapzip is fine, I just happened to have that patch floating around when someone mentioned it in the logs
16:32:56obobut it would be handy if the build system kept them around with the archived daily builds :)
16:33:08bluebrother*grrr* always this stupid web designers that require you to maximize the browser window
16:33:21B4gderah right, for daily builds it would be very easy to keep the maps around...
16:33:28*low_light couldn't resist a Sansa C250 for $15
16:33:42B4gderlow_light: another mi4 port for you!
16:34:11B4gder"just" an mrobe is too easy ;-P
16:34:16*bluebrother would be heavily tempted too
16:34:55amiconnlow_light: Btw, what's the status of your m:robe port?
16:35:02barrywardellamiconn: care to look over my PP502x sleep patch?
16:35:35 Join toffe82 [0] (
16:35:52bluebrotheroh, we'll get sleep mode?
16:36:12 Quit linuxstb ("Client Exiting")
16:36:24low_lightamiconn: still no touchpad
16:36:49barrywardellbluebrother: not yet. my patch just allows running the cpu at a very low frequency (32kHz). the same as is already available for PP5002
16:37:27amiconnbarrywardell: Sure, could you put it up somewhere?
16:37:42barrywardellhere it is:
16:37:42bluebrothertoo bad ... but if you need testing I could try on a PP5022B
16:37:58barrywardellquite a few whitespace changes in there just to keep things tidy
16:38:34barrywardellbluebrother: it would be good to get some testing on PP5022
16:38:51 Quit Entasis ("Leaving")
16:39:12bluebrotherjust post a patch somewhere and tell me what to try ;-)
16:40:11amiconnbarrywardell: Ah, you've changed it so we're always running from source #2 after exitiing the function...
16:40:40barrywardellyes, I tried to make it more like the PP5002 version, which I find is easier to understand
16:41:09 Join Febs_ [0] (
16:41:47amiconnI agree that it makes sense...
16:41:48barrywardellactually, source 2 was always the selected source
16:41:58amiconnNot for CPUFREQ_DEFAULT
16:42:24barrywardellwhy not?
16:43:11amiconnCPUFREQ_DEFAULT used to set use_pll = false
16:43:12barrywardellah, I see. nvm
16:43:28barrywardellbluebrother: the pastbin l
16:43:32barrywardellI linked has the patch
16:43:35amiconnBut that looked a bit kludgy, even though I added it ;)
16:44:10B4gderok, starting tomorrow there will be a maps zip for each daily build
16:44:29barrywardellbluebrother: you'll also need some way of setting the cpu frequency to the CPUFREQ_SLEEP and testing that it's actually working
16:44:34amiconnbarrywardell: I think that patch should be committed if it's confirmed stable on PP5020 and PP5022
16:44:38barrywardellamiconn: I agree
16:46:58barrywardellamiconn: I'll run it on my H10 for the rest of the day and do the same on my Sansa. That just leaves PP5022 to be tested
16:46:58 Quit DogBoy (Read error: 104 (Connection reset by peer))
16:47:15amiconnI'll put it on my mini g2...
16:47:53barrywardellcool, can you check that it's actually running at 32kHz too?
16:48:17amiconnShould be simple...
16:48:41*amiconn often just sticks such tests in debug_menu.c:dbg_hw_info()
16:48:54 Join BobShield [0] (
16:52:41 Join dandin1 [0] (
16:54:24oboB4gder: nice - would it be useful for current builds as well?
16:55:17 Quit jhMikeS (Read error: 104 (Connection reset by peer))
16:55:23B4gderI'd rather not, since it'll become a lot more transfers from all build servers
16:55:34B4gderlet's start daily and see
16:55:38 Quit kubiixaka (Read error: 110 (Connection timed out))
16:56:34bluebrotherbarrywardell: hmm, with only the patch applied I get a small gap in music output when entering the HW info screen.
16:56:36Soul-SlayerDidn't we agree r14099 caused some sort of bug? Or was that resolved? (The 'Add delete to WPS context menu' commit)
16:57:04 Quit Febs (Read error: 110 (Connection timed out))
16:57:34 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
16:58:21barrywardellbluebrother: hmmm...did you get that gap before the patch?
16:58:48bluebrothernot sure ... will try a clean build
16:59:09bluebrotherotherwise it plays fine so far.
16:59:53barrywardellideally, it will behave exactly as before the patch
17:02:29 Quit B4gder ("It is time to say MOOO")
17:03:00bluebrotherwouldn't it be NPPP this time? *g*
17:03:03amiconnbluebrother: That's normal, and it was there before
17:03:14amiconn(the gap)
17:03:27amiconnThis is because perfcheck() disables fiq
17:03:50bluebrotherok. I was just about testing
17:07:18 Quit BobShield (Read error: 104 (Connection reset by peer))
17:08:39 Join linuxstb [0] (i=d556da1b@rockbox/developer/linuxstb)
17:09:22 Join BobShield [0] (
17:09:23 Quit Febs_ (Read error: 110 (Connection timed out))
17:13:04***Saving seen data "./dancer.seen"
17:14:07 Quit linuxstb ("CGI:IRC (Ping timeout)")
17:14:56 Join Febs_ [0] (
17:14:58 Nick Febs_ is now known as Febs (
17:17:22 Join daurnimator [0] (n=fake@unaffiliated/daurnimator)
17:18:14amiconnbarrywardell: Sleep and clock setup in general are working fine on mini g2
17:18:31amiconnSleeping for 16384 loop cycles takes ~2 seconds
17:19:28amiconnStarted playback with repeat all enabled...
17:19:48barrywardellgreat :)
17:21:45*amiconn chose to use a C delay loop this time
17:22:12amiconnI know to what arm instructions it compiles. Just for(i = 16384; i > 0; i−−);
17:23:32barrywardellsame here on H10
17:24:59barrywardellit takes approximately the same time too
17:25:21amiconnYes, as expected
17:25:38amiconnThe PP5020 and PP5022 have equal performance at the same clock
17:26:14amiconnIt's just that PP5022 can be clocked higher (which we don't use), and supposedly draws less power at the same frequency
17:26:41amiconnPP5002 is slower from dram due to the broken cache. From iram it's as fast as PP502x
17:27:41barrywardellinteresting. the test does funny things to the Sansa lcd
17:28:36barrywardellit ends up with the origin somewhere near the middle of the screen after running it
17:29:41 Quit low_light ("CGI:IRC (Ping timeout)")
17:29:56barrywardellbut it does work as expected otherwise
17:30:27*amiconn leaves for a while
17:32:46 Quit ShadowXP (Read error: 104 (Connection reset by peer))
17:32:59 Join ShadowXP [0] (
17:36:53 Quit Febs (Read error: 110 (Connection timed out))
17:37:38 Join joshin_ [0] (n=joshin@VDSL-130-13-252-169.PHNX.QWEST.NET)
17:40:26 Quit joshin (Read error: 110 (Connection timed out))
17:48:55 Nick joshin_ is now known as joshin (n=joshin@unaffiliated/joshin)
17:56:12 Join joshin_ [0] (n=joshin@VDSL-130-13-9-132.PHNX.QWEST.NET)
17:57:52 Quit joshin (Nick collision from services.)
18:01:36 Quit toffe82 (Remote closed the connection)
18:10:39 Join Domonoky [0] (
18:10:46GodEater_after a day of frantic activity, it's all gone quiet
18:11:24pixelmabut your favourite e200R user posted again in the forums...
18:11:57*GodEater_ washes his hands of that thread
18:12:44GodEater_oh what a surprise - he's asking the same question all over again
18:14:16 Nick joshin_ is now known as joshin (n=joshin@unaffiliated/joshin)
18:16:05 Join Domonoky_ [0] (
18:16:09 Quit petur ("work->home")
18:21:44barrywardellhmmm, crash on my H10. When you get back, amiconn, here's an updated version that doesn't crash:
18:22:33 Join toffe82 [0] (
18:22:44barrywardellI had to move CLCD_CLOCK_SRC to after setting CLOCK_SOURCE or else i would crash when the lcd got re-enabled after sleeping
18:25:54toffe82Domonoky : can you make rbutil portable, what I mean is that actually all the downloaded files are kept in the user directory of the computer and it would be better to have everything in the same directory
18:26:17toffe82the same directory as rbutil
18:27:01toffe82and also, why keeping all the files ?
18:27:27Domonoky_toffe82: to which rbutil version are you referring now ? Qt or wx ? :-)
18:28:24toffe82I have a folder with 22mb of rockbox files on my user folder :)
18:28:30Domonoky_toffe82: if you use the clean download cache function, all those files are deleted...
18:28:39Domonoky_22mb :-)
18:29:19bluebrothermillibytes? ;-)
18:29:44 Quit obo ("bye")
18:29:53bluebrotherI was thinking of adding a configuration for the local repository to rbutilqt
18:30:05Domonoky_the files are kept for caching.. so if you install the same build again, it doesnt download it again.. ( doesnt really make sense, now that we recommend the current build and not dailies)
18:30:10toffe82I don't see the otpion on the 3.2.6
18:30:42Domonoky_toffe82: in the dropdown menü..
18:31:20Domonoky_where caching makes sense: is for the images in the theme installer..
18:31:33bluebrothertoffe82: from the "File" menu
18:32:26 Quit Domonoky (Read error: 110 (Connection timed out))
18:32:27toffe82ok, I miss this one, but what I said before for a portable version is still valid ;)
18:32:37bluebrotherbtw, the output of isn't really nice for parsing :(
18:32:41Domonoky_yes.. thats true..
18:33:17Domonoky_bluebrother: its not to bad, just do some crude hack, like i have done in the wx version :-)
18:33:46bluebrotherwell, as the themes web site is about to be replaced I think we could also adjust the php script to something that's better suited.
18:34:08bluebrothereasier to code, less potential errors ;-)
18:34:25Domonoky_bluebrother: thats true.. scorche, your job :-)
18:35:15bluebrotherini format or some simple xml like would be best I guess.
18:35:46bluebrotherscorche: if you give me the php file I can look into improving it ;-)
18:36:28bluebrotherbut I think he's out for the next week.
18:37:19 Join TickyTacky5 [0] (n=na@
18:37:33 Part Stalwart ("WeeChat 0.2.6-cvs")
18:38:48TickyTacky5im using windows xp and have a 30gb 5.5generation ipod with rockbox do i choose the default boot firmware? it defaults to rockbox, i would like it to default to the apple firmware
18:40:41bluebrotheryou can't do that.
18:40:46KasperleTickyTacky5: i don't know whether that is possible, let alone how to do it. but do you know that you can just put it on "hold" when rebooting, to boot the apple firmware?
18:40:54bluebrotherRockbox bootloader will _always_ default to Rockbox.
18:41:27bluebrotherbut if you boot into AppleOS and shut down the player goes into a sleep state. I.e. if you start the player from that state you will be in AppleOS anyway.
18:41:43bluebrotherto go back to Rockbox you need to reboot using a hardware reset.
18:41:49bluebrother... but the manual covers this.
18:41:51Kasperlei didn't even know you could shut down the apple os manually?
18:42:14bluebrotherwell, I can do on my mini. Hold Play for a couple of seconds.
18:42:36bluebrotherDon't know about newer Ipods but I don't think they removed that function ;-)
18:42:42TickyTacky5if i shutdown while in apple os, then turn it on, it boots to rockbox
18:43:03bluebrotherhow do you shutdown the player?
18:43:13TickyTacky5hold play/pause
18:43:23bluebrotherApple does *not* do a proper shutdown of the hardware so it can startup faster.
18:43:57bluebrotherthat's also the cause why booting Rockbox takes longer than booting AppleOS.
18:44:09bluebrother(well, booting in terms of "waking up")
18:44:23TickyTacky5its the other way around for me, rockbox boots fairly quick, but apple os takes a while
18:45:07bluebrotherwell, speaking of a non-rockboxed Ipod, AppleOS is already loaded.
18:45:31bluebrotherThe Ipod does a hard shutdown after a bunch of hours (I think it's at least a couple of days)
18:45:56bluebrotherin that case booting takes longer as it needs to load everything again.
18:46:43bluebrotherbut if you "shutdown" AppleOS and wake it up a short while later it will not reboot.
18:47:01bluebrotherand due to this up almost instantly.
18:47:06TickyTacky5also i saw someone playing pokemon using rockbox and i tried it and a few other gameboy roms which ive seen run on videos online but they were running horribly for me
18:48:02*bluebrother is not interested in playing games on the Ipod
18:48:43 Quit XavierGr (Nick collision from services.)
18:48:45 Join XavierGr_ [0] (
18:49:45 Quit TickyTacky5 ()
18:51:32 Quit ShadowXP (Read error: 104 (Connection reset by peer))
18:51:59 Join ShadowXP [0] (
18:56:10 Quit pondlife ("disconnected has pondlife")
18:57:02 Join a1titude [0] (
19:06:22 Join Xerion [0] (
19:07:42 Join robin0800 [0] (
19:08:03 Join lazka [0] (
19:13:06***Saving seen data "./dancer.seen"
19:18:19 Join petur [0] (n=petur@rockbox/developer/petur)
19:19:05 Quit petur (Remote closed the connection)
19:19:19 Join petur [0] (n=petur@rockbox/developer/petur)
19:19:23 Join IceGuest_7 [0] (
19:20:26 Nick IceGuest_7 is now known as Toxicity (
19:20:55 Join IceGuest_7 [0] (
19:21:24 Quit barrywardell (Remote closed the connection)
19:21:59 Quit ze (Remote closed the connection)
19:22:02 Join ze [0] (
19:25:53 Nick IceGuest_7 is now known as FCUK (
19:26:01FCUKfucking fucked up fucks
19:26:44FCUKnigga please
19:26:45FCUKnigga please
19:26:47 Join obo [0] (n=obo@rockbox/developer/obo)
19:26:47DBUGEnqueued KICK FCUK
19:26:48***Alert Mode level 1
19:26:48DBUGEnqueued KICK Toxicity
19:26:49***Alert Mode level 2
19:26:49***Alert Mode level 3
19:26:49***Alert Mode level 4
19:26:49***Alert Mode level 5
19:26:50***Alert Mode level 6
19:26:50***Alert Mode level 7
19:26:51***Alert Mode level 8
19:26:59***Alert Mode level 9
19:27:05***Alert Mode level 10
19:27:07***Alert Mode level 11
19:27:09***Alert Mode level 12
19:27:11***Alert Mode level 13
19:27:19***Alert Mode level 14
19:27:21***Alert Mode level 15
19:29:07 Join robin_0800 [0] (
19:29:31peturBagder ?
19:29:36***Alert Mode level 16
19:29:37***Alert Mode level 17
19:29:38***Alert Mode level 18
19:29:38***Alert Mode level 19
19:29:39***Alert Mode level 20
19:29:40***Alert Mode level 21
19:30:14peturamiconn ?
19:30:20 Join tomaw [0] (i=tom@freenode/staff/tomaw)
19:30:54 Quit hashier (Read error: 104 (Connection reset by peer))
19:30:58 Join hashier [0] (
19:31:15 Join XavierGr [0] (
19:31:23CtcpIgnored 2 channel CTCP requests in 1 second at the last flood
19:31:23*robin_0800 slaps FCUK's bottom and grins cheekily
19:31:41FCUKfuck you
19:33:03Torinyeah great, what an awesome level of culture
19:33:11FCUKfuck you
19:34:56 Quit robin0800 (Read error: 110 (Connection timed out))
19:34:57 Nick robin_0800 is now known as robin0800 (
19:35:00FCUKfuck you
19:36:06FCUK[12:29] <FCUK> enlarge
19:36:06FCUK[12:29] <FCUK> viagra
19:36:06***Alert Mode level 22
19:36:06FCUK[12:29] <FCUK> nigger
19:36:06***Alert Mode level 23
19:36:06FCUK[12:29] <FCUK> fuck
19:36:28 Quit FCUK (Excess Flood)
19:36:33 Quit Toxicity ("If your not living on the edge, you're taking up too much space")
19:36:35 Quit ShadowXP (Read error: 104 (Connection reset by peer))
19:36:48lazkawtf was that?
19:36:57petura troll
19:37:03 Join ShadowXP [0] (
19:37:21tomawSelf-fixing spambot :)
19:37:35bluebrotherbut he didn't include "buy now" links ...
19:37:53lazkawhat is the benefit of saying "F*CK N*GGER..." 1000 times?
19:37:59lazkai don't get it
19:38:14bluebrotheryou have something to do: typing 1000 times the same
19:43:46 Quit XavierGr_ (Read error: 110 (Connection timed out))
19:46:07***Alert Mode OFF
19:52:42 Quit ShadowXP (Read error: 104 (Connection reset by peer))
19:53:02 Join ShadowXP [0] (
19:54:43amiconnhmm, barrywardell is gone :/
19:54:46*petur tests if has a working anti-abuse service
19:54:56pixelmatomaw: ah, freenode staff watching over us?
19:55:18tomawonly because of the spambot earlier :)
19:55:30pixelmagood to know :)
19:55:45amiconnI think I know why Barry's 1st version crashed on H10 and how to avoid that a bit cheaper
19:57:20pixelmapetur: I believe that this is also Llorean's provider...
19:57:26 Part a1titude
20:01:59maddlerevening all
20:08:57 Join Llorean [0] (n=llorean@
20:13:01 Quit robin0800 (Read error: 110 (Connection timed out))
20:13:41amiconnWith sleep test that properly disables FIQ as well, even calling it while music is playing is no problem. Music just stops during the test, then continues
20:16:00*petur would have expected the buildtable to be gapless now
20:17:28*amiconn wonders why people care about that gap so much
20:17:55LloreanObsessive-compulsive disorder. It just looks wrong. :-P
20:18:29amiconnAh, hmm. Maybe some manual adjustment is necessary now
20:19:07LloreanI'd have to talk to Zagor to get something changed on Flyspray, right?
20:19:14*Llorean really wants the "Vote" feature disabled, at least for bugs.
20:21:22 Quit desowin ("use linux")
20:24:08tomawYou guys seem to be okay now. If there's a problem in the future prod someone listed by /stats p and we'll take a look in to it :)
20:24:12 Part tomaw
20:26:01LloreanWe had a problem?
20:26:41peturcheck the log, a bit after 19:00
20:26:50pixelmaa troll - joining from
20:30:12 Join jgarvey [0] (
20:35:15LloreanAh, that is 'special'
20:39:30 Join rasher [0] (n=rasher@rockbox/developer/rasher)
20:41:53rasherCould anyone help me figure out why crosscompiling the sim no longer works for me. I get "No rule to make target `SDL.h'" (needed by /path/button.o). How can I figure out why it breaks on SDL.h?
20:42:29rasherCompiling windows sim on Linux, that is. It's the same setup I've been using for ages now. I can't see what might have changed.
20:47:46preglowwow, eventful evening
21:00:33 Join kubiix [0] (
21:02:44rasherHrm, didn't we have long heated arguments about capitalisation once, and ended up deciding to use the rules of English, regardless of the whim of manufacturers? (ie. Ipod rather than iPod, Iaudio rather than iAudio)
21:03:59 Join david__ [0] (n=chatzill@
21:05:25pixelmaIf you refer to the changes in the wiki - I haven't been around for any discussions about it but I think it looks wrong now, too.
21:05:34 Join DervishD [0] (i=50232cd4@gateway/web/cgi-irc/
21:06:40rasherThat's where I saw it, and thought there were other uses in the manual as well, but it seems it's just a single (wrong) use of iPod in the installation instructions
21:08:06amiconnI'm not sure whether the discussion was about capitalisation in general
21:08:21rasherIt was probably mostly related to the manual, yes
21:08:23amiconnI only remember that it's Iriver instead of iRiver
21:08:38amiconnBut iAudio and iPod should be written like that, imho
21:08:46david__can someone tell me why the ipod firmware loads by default after i disconnect my ipod from my computer?
21:08:49LloreanIf I recall, it covered Iriver, Ipod, Iaudio, but was directed to the manual since that's the part we control.
21:09:34Lloreandavid__: If you're in the Apple firmware, you have to use Menu+Select to leave it.
21:10:04rasherIt's just the easiest to simply use the English rules. Saves having to decide between iRiver <-> iriver, iAudio <-> iAUDIO etc.
21:10:06david__right. that does a "hard reboot", right? i was just wondering if this is supposed to be happening
21:10:16david__thx, llorean
21:10:44rasherAnd subjectively, I think it looks better.
21:11:21rasherI guess it's a lost cause in the wiki though. I wonder why someone took the time to correct those two pages
21:11:22LloreanI think, if you say it as a word "Ipod, not I P O D" then you should capitalize it as a normal word. We aren't putting logos throughout our text, merely words.
21:12:09pixelmaI agree completely - I also think this way we are "safe" in case the companies decide to spell it differently. It's just a name.
21:12:19*Domonoky_ has commited nice little icons for the progress logger in the new rbutilqt :-)
21:12:47 Join lostlogic [0] (n=lostlogi@rockbox/developer/lostlogic)
21:12:53rasherAnyway, the manual already does this. I don't think further discussion is necessarry. Just popped into my head when I saw the wiki edits.
21:13:10***Saving seen data "./dancer.seen"
21:13:40lostlogiclatest svn doesn't build without db / dircache
21:13:48david__hey, one more little question. i'm getting clipping while listening to my database every 6 to 8 seconds, and i figured that maybe i can fix this by installing the latest build of rockbox. how do i upgrade? should i simply overwrite the .rockbox files with the new one?
21:13:50lostlogicdebug menu ifdefs are effed
21:14:40DervishDdavid_, yes, you have to do just that
21:14:57david__thx. any idea on that problem, thougH? have you ever heard of that?
21:15:16 Join ComposerDude [0] (i=0cdb6940@gateway/web/cgi-irc/
21:15:20david__every 6 to 8 seconds, the sound stops and then plays again after a few milliseconds
21:15:20DervishDI don't have that problem in my iAudio X5V
21:15:34DervishDWhich player do you have?
21:15:41david__damn. i don't think i had it until i started moving a lot of flac files onto the ipod, too
21:15:45david__Ipod 4th gen grayscale
21:15:46rasherdavid__, that's not what's called clipping. What you're hearing is the player failing to keep up with decoding the file.
21:15:52 Join haemmy [0] (i=stefan@
21:15:53 Quit lazka (Remote closed the connection)
21:16:02rasherdavid__, do you use the equalizer?
21:16:03david__well, sounds the same
21:16:12DervishDIt's weird, an iPod short of CPU :?
21:16:24 Join DerPapst [0] (
21:16:26DervishDSpecially with FLAC,
21:16:28david__no. i tried using that, and it would be almost impossible to use it, so i disabled the eq
21:16:58*DerPapst agrees with amiconn
21:16:58rasherIs this while listening to flac?
21:16:58DerPapst i think it should be iPod and iAudio too
21:17:01david__but i still get the disruption, like a buffer problem maybe, but ever few seconds
21:17:21david__actually, if i listen to the flac files i uploaded to the ipod's root directory (even in database mode), no problem
21:17:32DervishDThat's even weirder :?
21:17:33david__but if i listen to the files i previously had on the ipod, i get the constant interruption
21:17:39rasherdavid__, which files are these?
21:17:45DervishDHave you checked they're not damaged?
21:18:12rasherWell, if they're aacs, it's no wonder
21:18:26david__um. it's pretty much every mp3 file i already had on the ipod before i started loading it with flac files
21:18:34david__i suppose they could be damaged..?
21:18:44DervishDalmost all?
21:18:46rasherI'm fairly sure it's a cpu problem.
21:18:54rasherIt sounds exactly like it. How old is your build?
21:18:55david__i must have tried dozens. every one
21:19:00david__but it's only in rockbox firmware
21:19:05pixelmadavid__: are there peakmeters in your wps?
21:19:05DervishDAnother suggestion: do fsck.msdos
21:19:06david__it's working in the ipod firmware
21:19:23david__pixelma, i don't know what peakmeters or wps are
21:19:23 Quit crashd_ ("leaving")
21:19:30 Join crashd_ [0] (
21:19:33rasherdavid__, little bars moving with the music
21:19:45david__oh, yeah
21:19:50LloreanExactly. They can make their logo look however they want, we're simply spelling the word that people actually speak, instead.
21:19:50LloreanIs there a WikiGuidelines somewhere?
21:19:50LloreanOr something similar?
21:19:51pixelmawps = while playing screen
21:19:54Lloreandavid__: Yes, though clipping usually suggests you've got the settings wrong. Are you using the equalizer?
21:19:56david__yes, there are
21:20:00Lloreandavid__: What do you mean by "clipping"?
21:20:02LloreanAre you using it in the audio context, or do you mean "the sound cuts out every few seconds"?
21:20:19david__the sounds cuts every few seconds. i'm not sure what the actual source of this interruption is
21:20:23rasherLlorean, welcome back to the future.
21:20:25david__and i am not using the equalizer
21:20:25DervishDSorry for not helping more, but I have to go now (my wife's gonna kill me...)
21:20:26david__i know that much
21:20:32DervishDPleased to meet you :)
21:20:34david__oh, thx anyway
21:20:38 Quit ComposerDude ("CGI:IRC")
21:20:39 Quit DervishD ("CGI:IRC (EOF)")
21:20:48rasherdavid__, It's probably because of the peakmeters, try switching to a different WPS
21:20:56david__wps means..?
21:21:02rasherWhile Playing screen
21:21:09david__oh ok
21:21:12amiconnLlorean: Could you run a special build both on a nano with flash version 1.2.x and 1.3.x?
21:21:21rasherdavid__, switching themes will change the wps as well
21:21:52amiconnPreferably with -wf'ing rockbox.ipod to the firmware partiton, so the bootloader won't fiddle with the ata settings
21:21:54 Quit Llorean ("Leaving.")
21:21:58 Join Llorean [0] (n=llorean@
21:22:09*Llorean sighs
21:22:14david__i am using the default theme, btw
21:22:29david__but i will try to switch the you mean that i should accomplish this by switching themes?
21:22:34david__or that's in the settings
21:22:42Lloreanrasher: I'm at a hotel with incredibly shaky internet access. So when I say things doesn't necessarily align with when you guys see them, at all. ;)
21:22:49rasherdavid__, Just switch to a different theme
21:23:09rasherdavid__, you can change the wps seperately as well, but the easiest is to switch the theme
21:23:28david__i haven't even explored the themes yet..
21:23:36david__i see them in the settings, thoughh
21:23:53rasherI don't really know what's a good suggestion to use
21:23:59david__do i have to reboot for the theme to change?
21:24:05david__after setting it to another
21:24:14rasherIt works instantly
21:24:40amiconnLlorean: saw my question?
21:24:58DerPapstinstand rockbox! now new with chocolate flavour
21:25:27Lloreanamiconn: Missed it, found it in the logs. Assuming the files DerPapst gave me will allow me to downgrade my flash it should be no problem.
21:25:41LloreanThough I'll try in 1.2 first, actually, so it's no problem either way
21:25:45DerPapstthis will work
21:25:51DerPapsti'm 99% sure about it
21:26:03LloreanYeah, I'm 99% as well, but there's always that sneaky little 1% ;)
21:26:15LloreanI never trust *anything* until I've done it myself.
21:26:30pixelmaif DerPapst himself says that who will doubt ;)
21:26:42DerPapstwell it worked on other iPods so far. Including my 5.5G iPod... so i would wonder if it doesn't work
21:27:04DerPapstDerPapst == Unfehlbar :D
21:27:10 Quit netmasta10bt (Read error: 110 (Connection timed out))
21:27:40 Join sharpe [0] (
21:27:43david__i keep trying to change the theme, but it just looks the same
21:27:56LloreanDerPapst: Well, I can't even make my flash dump in Rockbox right now. For all I know, mine has exploded and that's why my iPod is the last one working. :)
21:28:08Lloreanamiconn: Is this special build posted somewhere, or are you going to just give me a diff?
21:28:11pixelmadavid__: did you install the
21:28:31rasherdavid__, you probably don't have the extra fonts installed. They can be downloaded on the current build page
21:28:43david__i should probably just upgrade to the current build anyway
21:28:43amiconnLlorean: A diff. It's a hack that reads the ata register range (in main.c) and then dumps it after all the inits are done
21:28:53DerPapstLlorean: what exploded?
21:29:28DerPapst(sounds scary)
21:30:03LloreanDerPapst: My flash ROM? Who knows?
21:30:24Lloreanamiconn: Alright, is there a link in the logs that I missed, or are you writing it now?
21:30:27amiconnLlorean: Hmm, I wonder whether this dump will work then...
21:30:31pixelmadavid__: the fonts are in a separate zip you should install additionally (the manual has links and explanation)
21:30:48david__oh, ok
21:33:14LloreanCompiling now, what am I looking for. The dump is automatic on boot?
21:33:33LloreanYou'll just need the ata_reg_dump.bin files?
21:35:28rasherHrm.. rbutil doesn't fit on a standard GNOME desktop in 800x600 resolution
21:35:56GodEaterI blame the UI designers
21:36:06DerPapstI really don't get why your flash rom should be faulty... I mean the iPod still boots and discmode and diaconstic mode are working too. So if there is something wrong with it i doubt your iPod be working. I would more think of a strange bug in rockbox.
21:36:34GodEaterDerPapst: I think llorean was kidding when he said he thought it had exploded
21:36:36Lloreanamiconn: 512 bytes a reasonable size?
21:36:57DerPapstGodEater: anywas... he can't dump it
21:37:08DerPapstso something is wrong then
21:37:20DerPapstand dumping works at least on my 5.5G iPod
21:37:25GodEaterperhaps it's the wrong day of the week :)
21:37:44DerPapstalready 2 days ago huh? ;)
21:37:56amiconnLlorean: DId you do it with 1.2 or 1.3?
21:38:20Lloreanamiconn: 1.2
21:38:23LloreanFlashing 1.3.1 now
21:39:01GodEaterthere's a 1.3.1 out there now ?
21:39:20DerPapstLlorean: btw have you used the bootloader for this test or have you installed rockbox.ipod on the firmware partition?
21:39:28LloreanDerPapst: Bootloader
21:39:33DerPapstGodEater: for nanos yes.
21:39:56DerPapstLlorean: because amiconn said to write the rockbox.ipod to the fw partiton
21:40:01LloreanAh, missed that
21:40:02DerPapstwithout using the bootloader
21:40:11LloreanI'll do that then
21:40:19DerPapstso that it doesn't mess with whatever rockbox dumps now
21:40:31DerPapstok ;)
21:40:51amiconnThe bootloader also calls ata_init() which changes some of the registers I'm interested in
21:40:58 Quit The-Compiler (Read error: 104 (Connection reset by peer))
21:41:20amiconnThe hack reads the registers before ata_init() is called in main rockbox
21:42:36LloreanAlright, well I have a good 1.3.1 dump, and I'll redo that 1.2 dump in just a minute.
21:43:59 Join The-Compiler [0] (
21:44:05 Quit rasher (Remote closed the connection)
21:45:15 Quit The-Compiler (Client Quit)
21:45:31LloreanI'll fix
21:45:58LloreanTry again?
21:46:49amiconnthx"> should be there now too
21:47:41 Join robin0800 [0] (
21:48:02*DerPapst sees ... numbers... and letters like 11 and QT :P
21:49:10*DerPapst is curious how Llorean's iPod behaves now. Will it downgrade or net... ;)
21:49:17 Join rasher [0] (
21:50:19 Join AceNik [0] (n=AceNik@
21:50:20amiconnLlorean: Thanks.
21:50:32AceNikuys is the server down, i cant update?
21:50:48AceNikit says unknown hostname
21:50:54GodEaterthey appear identical to me amiconn
21:50:57LloreanDerPapst: Music is playing without data aborts.
21:50:57amiconnUnfortunately the register setup is identical
21:51:29amiconnBut the setup is rather different from what rockbox does
21:51:32GodEaterAceNik: we're not very good at guess the hostname
21:51:33DerPapstLlorean: ooh?
21:51:35amiconnHave to investigate...
21:51:35LloreanWell, let me upgrade to 1.3.1 again, and see if it breaks playback, since that's not what I tested.
21:51:42LloreanDerPapst: That just means I downgraded to 1.2 safely
21:51:46DerPapstLlorean: can you dump the flash now btw?
21:52:19*amiconn wonders when his 2nd gen's battery will give up...
21:52:26AceNikunknown hostname ""
21:52:28amiconnIt was playing all day without recharging...
21:52:31GodEateramiconn: is this your aftermarket one ?
21:52:56amiconnVoltage is still ~3.65V
21:52:57GodEaterAceNik: unknown hostname does not mean it's down, it means your DNS doesn't know it
21:53:16AceNikwhat do i do , it just stopped suddenly ? has address
21:53:25amiconnGodEater: It's a NewerTech 1600mAh battery
21:53:43GodEaterAceNik: you could add that to your host file
21:53:53GodEateramiconn: what's the stock mAh on a 2G ?
21:53:54amiconnStock battery is 1230mAh according to
21:53:57 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
21:54:00GodEateryou read my mind
21:54:09AceNikgodeater: i was tryin to do a svn up , while my net connetion switched off, after i reconnect it doesnt update, & gives this error
21:54:13GodEaterhow long have you got out of it so far
21:54:28 Quit haemmy ()
21:54:42GodEaterAceNik: as I said, add it to your hosts files
21:55:44amiconnGodEater: Umm, not sure. Runtime isn't correct because it (1) doesn't detect the charger and (2) I reset several times to put something on disk
21:56:10LloreanDerPapst: I'm not sure my player is upgrading to 1.3.1 properly
21:56:11amiconn(and no, it doesn't recharge when I connect it to my laptop's firewire connector, which is 4-pin)
21:56:11GodEaterearly days for a battery benchmark I guess
21:56:16LloreanAt least, it still says "1.2" in the about screen
21:56:39GodEaterLlorean is the owner of the worlds only unflashable nano
21:56:43amiconnI think it was playing for at least 8 hours today
21:56:54AceNikwhere is this s/files/file
21:56:57GodEateramiconn: that's pretty impressive
21:57:06GodEaterAceNik: depends on your OS - google for it
21:57:07LloreanGodEater: Yeah, but that's the on-disk firmware
21:57:38GodEaterLlorean: neat :)
21:57:46DerPapstLlorean: have you used the image i gave you or the once downloaded from
21:57:46LloreanOh, wait
21:57:49GodEateryou've set the whole device read only ;)
21:57:49Llorean1.3.1 is installed
21:57:53LloreanDerPapst: I found the problem
21:58:01DerPapstok ;)
21:58:03LloreanWhen booting into the OF it was loading the apple_os.ipod off my fat32 partition
21:58:10LloreanSo, the flash was fine, it just wasn't loading the right actual firmware
21:58:19LloreanWhich wouldn't have affected the test anyway
21:58:56*amiconn dives into OF disassemblies looking for accesses to the ata register range
21:59:10LloreanMost people said problems crop up within 5-7 seconds.
21:59:13LloreanMy Rockbox is working. :(
21:59:50amiconnMaybe it doesn't depend on the flash version, but on the installed flash disk chips
22:00:22amiconn...which might correlate to the flash version installed during manufacturing
22:00:27LloreanVery possible
22:00:50amiconnWhat does 'View Disk Info' display on the nano, btw?
22:00:56amiconn(disk type etc)
22:00:57LloreanIt would explain why it's a minority that's experiencing it, since it would be probably the last batch or two of 1st gen Nanos
22:01:31AceNikGodeater: how do i do a svn cleanup
22:01:45GodEaterAceNik: a what ?
22:01:51amiconnThe interesting details would be Model, Firmware, PIO modes and Cycle times
22:02:06AceNikyou know when there is a lock created in svn
22:02:22amiconnOh, and IORDY support and IORDY disable
22:02:42Lloreanamiconn: Model: SST55LD019K-45-C-MWE (may extend off the screen). Firmware : ADBA40KA. PIO modes: 0 1 2 3 4, Cycle times: 120ns/120ns
22:02:54LloreanIRODY support: yes, disable: no
22:03:09amiconnAnd now the same from a problematic nano...
22:04:08GodEaterAceNik: I have no idea - again, google for it
22:04:29 Quit AceNik ("bye guys have fun , enjoy !!!!!")
22:04:54Lloreanamiconn: Okay, requesting
22:06:00GodEaterdon't hold your breath
22:07:16GodEaterLlorean: "now problems" or "no problems" ? :)
22:07:53LloreanI don't know what you're talking about! ;)
22:08:05GodEaterSuuuure you don't (/me notices the edit)
22:08:10 Join miepchen^schlaf [0] (
22:08:48 Quit jhulst (Remote closed the connection)
22:09:02LloreanWhat edit?
22:11:08LloreanHopefully a few of them have notifications set on that thread.
22:11:45GodEaterlooks like I have to teach myself the menu api for my delete option
22:11:51GodEaternot tonight - too tired
22:15:53 Quit ShadowXP (Read error: 104 (Connection reset by peer))
22:16:05 Join ShadowXP [0] (
22:16:39 Quit XavierGr (Read error: 110 (Connection timed out))
22:26:29 Join barrywardell [0] (
22:28:21bluebrotherDomonoky_: just came back and noticed your last commit. Nice.
22:28:31Lloreanbluebrother: In regard to the Nano issue, we've got at least 4 people who can reproduce it.
22:28:52LloreanThough we haven't identified a correlating factor yet (it used to seem like it was firmware version 1.3/1.3.1 vs 1.2 and earlier, but I seem to have disproved that)
22:29:16bluebrotherI added icons to the Ok / Cancel buttons (and a few others too), but I need to separate that from the rest of my changes first.
22:29:31bluebrotherLlorean: why should the OF make a difference? It isn't even running ...
22:29:51Lloreanbluebrother: Hardware init in the flash part of it.
22:30:24bluebrotherdoes the firmware update of Apple also replace the flash?
22:30:39LloreanThat's why it'll insist you plug in the player during a portion of it
22:30:56bluebrotherhmm, never used itunes with my Ipod ;-)
22:31:04amiconnIt doesn't do so on all ipods afaik
22:31:20LloreanBut it at least does on 5G/Nano
22:31:34LloreanThe only permanent brick I've ever heard of resulted in someone holding Menu+Select during the progress bar of the flash update.
22:31:35amiconnBut on my 2nd gen the apple updater also initiated a flash update when I used it to convert it to a winpod
22:32:05amiconnMenu+Select is hardware reset, and it's not wise to reset while flashing...
22:32:46 Quit david__ (Remote closed the connection)
22:32:47amiconnIt's funny that the apple loader insists on plugging power for flashing, even if the battery is fully charged
22:32:59Soul-SlayerBetter safe than sorry
22:33:20amiconnThat irritated me when doing the conversion... it just displayed a firewire plug and an arrow, then shutdown...
22:33:37amiconnTook me a while to understand that the plug means it wants power...
22:33:56LloreanThough unlike it suggests, it doesn't have to be wall-power
22:34:10peturhmmm somebody should make the users clear that we're supposed to clean up the tracker, not add new items :)
22:34:42bluebrothernext time I add a "do NOT add new tasks during that period" notice to the announcement ;-)
22:35:01*amiconn wonders how far the battery voltage will go down until it dies.
22:35:18amiconnI expected ~3.5V, but it's at 3.3V now and still playing
22:35:41Domonoky_bluebrother: i also have the talkfile creator ported, but i dont have the ui for it at moment, and its to late for today :-)
22:35:50bluebrotherbarrywardell: I had my mini running with your patch and it worked nicely so far.
22:36:11barrywardellbluebrother: great. my sansa is still running fine too
22:36:12amiconnBut now I can see that backlight intensity changes when boosting/unboosting
22:36:22bluebrotherDomonoky_: I wanted to finish the theme installer but I had some RL stuff to do.
22:36:34barrywardelli'm just catching up on the logs. it seems amiconn has an idea about the h10 problem?
22:36:37bluebrotheralso, I think it's better to adjust rbutil.php first to make it less error-prone
22:36:46barrywardellit works fine with the change I made btw.
22:37:04bluebrotherbut I'll keep working on it ;-)
22:37:12amiconnbarrywardell: You're correct that the order of switching cpu and lcd clock matters. I thought my idea would be more efficient, but it isn't
22:37:25amiconnDoes your second version work without crashing?
22:37:28barrywardellah, ok.
22:37:43Domonoky_bluebrother: :-) not long, and we have it on pair with the wx version :-)
22:37:47barrywardellyes, the crash was very reproducable before
22:38:03bluebrotherDomonoky_: btw, what icons did you use? All other icons are taken from the tango project afaik
22:38:04barrywardellnot I haven't been able to get it to crash since and have had it running for several hours
22:38:08barrywardelluntil the battery died
22:38:17bluebrotherso checking for tango icons might look a bit more consistent ;-)
22:38:21amiconnWorks like a charm on mini 2nd gen as well
22:38:33Domonoky_bluebrother: the icons where from gnome-icon-theme
22:38:34amiconnI'd say so
22:39:12amiconnjhMikeS: around?
22:39:16Domonoky_bluebrother: feel free to change the icons, i wanted to get the code for it in.. :-) because you need those log calls everywhere..
22:39:55bluebrotherok, great.
22:40:22bluebrothergtg soon, but I think I'll give it a try soon, and also add my button icons.
22:40:57bluebrotherhehe, they even have icons for the Ipod. Now if we have icons for the other brands we could add icons in the device selection ;-)
22:41:16LloreanAlright, my Nano's still playing, which I think means officially that I cannot reproduce this issue. :(
22:41:19Domonoky_:-) we have small icon for the archos..
22:41:24rasherbluebrother, You could check the tango-artists mailinglist and request
22:42:00rasherSince we already have svgs from the manual, they should have an easy job
22:42:23bluebrothergood point.
22:42:46bluebrotherbut I gtg now. bbl.
22:42:53rasherTangofying existing svgs shouldn't be hard
22:43:02rasherI could write a mail if you need those icons?
22:44:46bluebrotherwell, I don't _need_ them. But it could make things a bit nicer.
22:45:10rasherAre the current svgs gpl or gfdl?
22:45:12bluebrotherbtw, we don't have a svg for Ipod 1G / 2G. pixelma ...?
22:45:31bluebrothernot sure. The manual is gfdl, so I guess the svgs are too.
22:45:45bluebrotherbut I'm out now. cu later.
22:45:48pixelmayes, will do one as soon as I get a decent scan... :)
22:46:14 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:46:28 Quit amiconn (Nick collision from services.)
22:46:30 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:46:45rasherHow many people besides pixelma have contributed device svgs, and would it be troublesome to relicense them under gpl if tango artists want to use them as a base for an icon?
22:46:58rasherDual-license, I suppose.
22:47:28rasherSince the icons will need to be gpl if they're to be included in rbutilqt
22:47:41LloreanAre they loaded from disk or compiled in?
22:48:07rasherCompiled in. Iirc, the goal is to pretty much have a single binary and a config file
22:48:29pixelmaas far as I'm aware 1 person helped - someone with the nickname "yossa" (Piotr...) helped with the initial drawing of the Sansa
22:48:53rasherpixelma, and you hold copyright on the rest of the svgs?
22:49:14 Quit Jens ("Oh yea...")
22:49:52 Part Domonoky_
22:49:52pixelmaif I'm not forgetting one...
22:50:24barrywardellI think I have a contact address for the guy that made the sansa svg...
22:50:43amiconnOkay, going below 3.4V on 2nd gen isn't recommended. Boosting is visible in backlight (gets darker) and lcd contrast drifts away
22:51:03pixelmabarrywardell: that's not the one you contacted last time
22:51:24barrywardellah, ok.
22:52:01 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
22:53:27pixelmaI should have his email address
22:54:06LloreanBagder: It's possible I can get access to an e200R to run tests on it, assuming they're all read-only tests (no writing as it's a friends and she may not entirely trust me unless I give her a good deal of reassurance that nothing will be changed)
22:54:11LloreanIf you had any specific ideas on tests to run
22:54:11rasherHrm, a Frederik Vestre created nano,mini and video svgs. But I'm not sure if they were replaced, or modified by pixelma
22:54:26pixelmathey were replaced
22:54:40linuxstbLlorean: Regarding flash upgrades and power - my ipod Color refuses to reflash when connected to USB and demands the wall adapter.
22:54:45BagderLlorean: I would expect some coding of test programs needed
22:55:05Bagderlike one that reads the entire NAND
22:55:08Lloreanlinuxstb: Mine asked for wall power, I plugged it into my laptop, the progress bar progressed like normal, and everything seemed to be upgraded.
22:55:10Bagderand scans for the BL
22:55:14LloreanSince I can't dump the flash, I can't be 100% sure though
22:55:20linuxstbBagder: Has anyone physically dissected an E200R to see if any components are different?
22:55:27Bagdernot that I know of
22:55:38linuxstbMaybe that's something to ask on the forum thread?
22:55:50rasherpixelma, okay, then you do hold all the copyrights except the sansa. I'm assuming you don't have a problem with using them under the GPL as well as the GFDL?
22:55:54Bagdermy dev board had Rhapsody sw installed and was identical to my "regular" ones
22:56:08Bagderso I've just pretty much assumed they're next to identical
22:56:40Bagderalthough this dev sw is a hybrid Rhapsody and not the same as in the actual R models
22:57:07linuxstbI'm just searching for places the bootloader could be... Another check would be the exact size of the flash disk - i.e. if the usb disk mode code is hiding some of it.
22:57:25Bagderlinuxstb: exactly what I just asked in the R thread ;-)
22:57:42 Quit Xerion (Read error: 104 (Connection reset by peer))
22:57:46pixelmarasher: not at all
22:58:54linuxstbBagder: So I see - good idea ;)
22:59:06rasherpixelma, Excellent. Maybe we could get some tangofied device icons then
22:59:38*Llorean wonders what exactly "tangofied" means.
22:59:49LloreanI'm aware of the Tango project, but what would it mean?
22:59:56amiconnAnyone else around who knows how to use the PP emulator?
23:00:35rasherLlorean, That they look like Tango icons. The same aspect, have a contrast-line, use the same colourscheme where applicable etc.
23:00:46LloreanSo, integration.
23:01:12rasherYeah, but the way the rules are set up means they look nice pretty much wherever you use the
23:01:30rasherSo they'll fit in on Windows, mac and Linux alike
23:01:37*amiconn thinks it's a funny coincidence that PP5002 is called Tango digital media platform...
23:02:05Bagderall part of the same conspiracy!
23:03:04pixelmarasher: hmm... can't find the old email of him directly (but I know he's registered in flyspray). And I'm almost sure he said "use as you like" but better to ask him again
23:03:34*linuxstb reads the latest post to the E200R thread and bangs head on table
23:04:01 Quit davina (Remote closed the connection)
23:04:02pixelmahe's also registered in the forums (same nick)
23:04:56rasherpixelma, I'll just send off the mail and just leave out the sansa svg. If we end up needing it, I'll see about contacting him
23:06:25 Quit atsea- (Read error: 104 (Connection reset by peer))
23:06:53 Join davina [0] (
23:10:08 Join Jens [0] (i=Jens@pdpc/supporter/active/Jens)
23:11:25 Join NirIzr [0] (
23:12:00NirIzrI was thinking, how hard will it be to convert an application as a plugin for rockbox?
23:12:50Bagderwoo, that's a neat data sheet
23:13:10LloreanNirIzr: That depends an awful lot on what the program is, how it was coded, what it requires, and other factors.
23:13:11***Saving seen data "./dancer.seen"
23:13:16LloreanBagder: "neat" in a good way?
23:13:18RondomNirIzr: depends on the applicati+#n
23:13:23Bagder285 pages of goodies, yes
23:13:32 Quit Jens (Client Quit)
23:13:37LloreanSo, the useful sort of data sheet?
23:13:52Bagdervery much so
23:14:04 Quit Rondom ("Ex-Chat")
23:14:50pixelmarasher: the only thing about those svgs - they contain a workaround for a problem that occured with a "clean" version (for objects with a complete "hole" in it). Don't know what the Tango guys need...
23:14:57barrywardelllinuxstb: did you see my comment on
23:15:02Bagderpins, register maps, the lot
23:16:01pixelmarasher: that problem occured when converting the svg to png/pdf which is used later
23:17:21rasherpixelma, Ah, well. I don't think there'll be any issues
23:18:22*amiconn also wants a register map - for pp :(
23:19:34Bagdereven the usb controller is in there
23:19:57 Quit bdgraue (Remote closed the connection)
23:20:03Bagderand LCD...
23:20:26 Quit Kasperle (Remote closed the connection)
23:20:26*petur looks for austriancoder... and fails :/
23:20:33pixelmarasher: basically I cut these objects into two when necessary, but I think they could be merged easily and they are not too many...
23:22:32Bagderso, you can actually load code into the TCC over usb by fiddling some pins on boot
23:24:32 Join bdgraue [0] (
23:24:57peturanybody know why recordings on RTC targets start with R and on non-RTC with rec_ ?
23:25:17Bagderpossibly to keep the name short?
23:26:25pixelmaon non-RTC targets the filename will just count from rec_0001 up - the filename of RTC targets contains a timestamp
23:27:51 Quit ShadowXP (Read error: 104 (Connection reset by peer))
23:27:54peturthat I know... ;)
23:28:06 Join ShadowXP [0] (
23:29:44 Join webguest44 [0] (i=4a841a2a@gateway/web/cgi-irc/
23:30:04linuxstbbarrywardell: Just read it now. I think my patch did more than your's - it removed all code from IRAM (the change in config.h), not just the ASM functions.
23:30:45 Quit petur ("*plop*")
23:31:24webguest44Hi. I was thinking about getting an ipod mini, but does rockbox support fm transmitters and stuff?
23:31:37amiconnLlorean: If I am right we're setting a far more aggressive pio timing than we should for the clock frequencies we use...
23:32:08amiconn(and we're using pio because we don't know how to use dma)
23:32:24barrywardelllinuxstb: I see. so do you think we need to test in all areas of rockbox, not just flac?
23:32:57barrywardellI can't find ICODE_SECTION defined anywhere else in the rockbox code
23:33:58barrywardellah, but there's ICODE_ATTR too
23:34:43rasherwebguest44, Depends on what type of transmitter it is. If it interfaces with the ipod in any way other than capturing the sound output, the answer is no.
23:34:50 Quit davina (Remote closed the connection)
23:35:09Lloreanamiconn: So the solution is either less aggressive, or figure out DMA?
23:35:13amiconnNow if our LA wizard were here...
23:35:27*amiconn would like to see some measurements of this
23:36:42 Join desowin [0] (
23:36:47 Join davina [0] (
23:37:50rasherI'll try asking again, since it seems few people were here last time I asked. Could anyone help me figure out why crosscompiling the sim no longer works for me. I get "No rule to make target `SDL.h'" (needed by /path/button.o). How can I figure out why it breaks on SDL.h?
23:38:07rasherMy build environment has worked previously, and I don't believe I've changed anything that should matter.
23:38:26 Join BigMac [0] (
23:39:18Bagderrasher: it looks like it gets a dependency on that header file but the header file is then missing... ?
23:39:36rasherBagder, how do I figure out why it's not looking where that header file is?
23:39:58Bagderrasher: start by browsing the generated dependency file
23:40:13Bagderdep-* something
23:40:35rasherAh, sim/dep-sim
23:41:26linuxstbbarrywardell: I've never managed to understand what effect IRAM usage has on the PP5020... In some cases when I've tried putting code in IRAM, it runs slower...
23:42:13linuxstbIt's very likely related to where the data is as well - ARM can use relative addresses to access the data if it's near the code.
23:42:28 Quit freqmod (Read error: 104 (Connection reset by peer))
23:42:38rasherBagder, hrm yeah, that just lists "/long/path/button.o \\n SDL.h"
23:42:51rasherI assume there ought to be a path in front of the SDL.h
23:43:15barrywardelllinuxstb: so code+data in iram is probably fast, but code in iram, data in dram is slower?
23:43:28linuxstbbarrywardell: Possibly...
23:43:36Bagderrasher: compare to a non cross-compiled build and see if you understand more
23:44:20amiconnlinuxstb: Did you compare the generated asm?
23:45:42amiconnCould also be cache aliasing effects - moving the code to iram causes the data to be at a different location in dram
23:45:44webguest44rasher: How about a fm transmitter? Oh wait thats capturing sound. darn. thanks anyway.
23:46:31rasherwebguest44, if it's controllable from the ipod, it won't work. If it simply plugs into the headphone output, it will
23:47:11rasherBagder, yeah that gets a real path. How is this dep file generated?
23:47:28Bagderrasher: tools/
23:47:40Bagderit's not an easy read though
23:48:06Bagderit uses gcc -MG -MM -MT basically
23:48:23Bagderit probably needs the include path for SDL.h in CFLAGS or something
23:49:06rasherI really wonder why it stopped working. Any changes recently, or is it likely to be a change at my end?
23:49:24BagderI can't remember any change in this area for a loong while
23:53:26 Join rotator [0] (n=e@rockbox/developer/rotator)
23:53:29DerPapstLlorean: about flash upgrades and power... afaik all iPod that can be connected over firewire request firewire power for updating. So the only iPod that are happy with usb power wile updating the flash are nano and video iPods. Afair the other nanos don't even accept usb wall chargers...
23:53:40 Part NirIzr
23:53:54LloreanDerPapst: The 2G Nano's can't be charged from the wall?!
23:54:11DerPapstwith a usb wall charger sure.
23:54:24LloreanWell, that's the only iPod charger I've seen sold these days, the USB one.
23:54:32DerPapstthey all can except 1G to 3G iPods
23:54:41*Llorean uses a powered USB hub.
23:55:00LloreanI've got too many things to charge, and four-port hub was cheaper than a single-port iPod charger.
23:55:02 Join hannesd [0] (
23:55:23webguest44rasher: I think it isn't controllable from the ipod because it plugs into the port at the bottom. it allowsd you to listen to the radio
23:55:33Soul-Slayerwebguest44: It won't work
23:55:42 Quit hannesd (Connection reset by peer)
23:55:43DerPapstdon't know but i tried to help someone who had updated his iPod and it refused to flash the flash connected to an usb wallcharger
23:56:00 Join hannesd [0] (
23:56:02 Quit hannesd (Read error: 104 (Connection reset by peer))
23:56:07webguest44ok....ok... i guess thats what the OF is for, then.
23:56:14Soul-Slayerwebguest44: Well, take a look here
23:56:24Soul-SlayerA list of all the accessories and whether or not they work
23:56:34 Join hannesd [0] (
23:56:51 Quit hannesd (Success)
23:57:20 Quit Thundercloud (Remote closed the connection)
23:58:13 Join Thundercloud [0] (
23:58:53 Quit desowin ("use linux")

Previous day | Next day