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

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

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

#rockbox log for 2008-02-23

00:00:49gevaertsLars_G: The two main ways seem to be "-DUSE_ROCKBOX_USB" in the EXTRA_DEFINES in the makefile, or #define USE_ROCKBOX_USB in firmware/export/config-<your-target>.h
00:00:50 Quit toffe82 (Read error: 104 (Connection reset by peer))
00:01:14Lars_Ggevaerts: meh I was doing it autoconf style, thanks a lot.
00:01:49 Join toffe82_ [0] (
00:01:51crzyboysterI was trying to install vmware on my computer by following these directions > I have downloaded both the installer and the image and am currently extracting it using 7zip. What should I do next? And what does "md5sum: b53ee439589acd7233d2a59752fbe759" mean? The wiki isn't really clear on where to put the image?
00:01:57 Nick toffe82_ is now known as toffe82 (
00:02:35Lars_GHmmmm whose vendor and device IDs are we spoofing for the stack?
00:02:57gevaertsLars_G: those of the OF for the same device.
00:03:08bluebrothercrzyboyster: an md5sum is a cryptographic has. Usually used to verify the integrity of a file
00:03:10Lars_GAh nice
00:03:53 Quit Domonoky (Read error: 104 (Connection reset by peer))
00:04:39 Join robin0800 [0] (
00:04:51crzyboysterShould I care about that? And what should I do with this file that's being extracted from the image > rockbox-flat.vmdk ? Should I install vmware first?
00:05:18crzyboysterMy bad, should've installed vmware first thing.
00:05:35 Quit desowin ()
00:05:42Bagdercrzyboyster: you can install that now and just make it use that vmdk file
00:06:51crzyboysterCan I change the location of the Rockbox.vmx , rockbox.vmdk , and rockbox-flat.vmdk files?
00:07:59crzyboysterSo every time I want to compile somthing, I should launch the rockbox.vmx file?
00:08:22Bagderwell, you should launch vmware, yes
00:08:46Bagderor install cygwin, or just go with linux all the way!
00:09:16crzyboystercygwin doesn't work on my computer for some reason...
00:09:30bluebrotherinstall linux ;-)
00:09:48 Join fasmaie [0] (
00:10:19*linuxstb 3rds
00:10:32crzyboysterI've never used linux in my life and have no patience to learn how to. And does Ubuntu have a full linux install included?
00:10:56*bluebrother is always annoyed by the requirement to develop on windows at work
00:11:04bluebrotherubuntu _is_ a linux distribution.
00:11:52crzyboysterIs there a way to dual boot Windows XP and Ubuntu? And will this "way" not end up destroying my beloved 6 year old computer?
00:11:52Lars_GUsing the new stack!! weee
00:11:58bluebrotheryou need to bring some patience if you want to program on Rockbox anyway.
00:12:11advcomp2019crzyboyster, yea there is a way
00:12:17bluebrothercrzyboyster: linux is an OS, not a computer-destroying-virus.
00:12:35Lars_GI will try it out strongly with OS X, the OF usb has been failing horribly with OS X
00:12:37bluebrotherand I use dual-boot since ... well, at least 10 years or so.
00:13:11crzyboysterI know... and I don't plan on programming on rockbox, just compiling my own build maybe and generating a lang file or two.
00:13:57 Quit ze (
00:13:57 Quit pabs (
00:13:57 Quit [omni] (
00:15:37gevaertsamiconn: I added disk->usb serial number support by encoding each character in two hex digits. I think I did everything right, but I obviously couldn't test.
00:15:50magmaniacpetur: you might already know this, but I found something reassuring in the CF spec: "For CF Storage devices, recovery from sleep mode is accomplished by simply issuing another command to the device. A hardware or software reset is not required." ( in "ATA Functional Differences") - so skipping the soft reset is okayed by the spec! (then again, the spec is constantly violated...)
00:16:09peturmagmaniac: yes I know
00:16:49peturthe problem is unifying the code - we can't use an ifdef like for ipodnano
00:17:20magmaniacright, that's why I was thinking about how to tell CF and HD apart
00:17:34peturI thought so ;)
00:17:36 Quit crzyboyster ("CGI:IRC")
00:17:48NJoin[omni] [0] (
00:18:11magmaniacbut with all these spec violations it is almost impossible
00:19:29NJoinpabs [0] (
00:22:18NJoinze [0] (
00:22:18 Quit toffe82 (Read error: 104 (Connection reset by peer))
00:22:39 Join toffe82_ [0] (
00:22:47 Nick toffe82_ is now known as toffe82 (
00:22:59 Quit Zarggg (Connection reset by peer)
00:23:02 Quit bluebrother ("leaving")
00:24:03Bagderstrangely short article in Norwegian
00:24:35rasherWow, that is short indeed
00:24:35Bagdersays "1 february"
00:24:53Bagderand we've got >800 referrers from it this month
00:26:59 Quit petur ("Zzzz")
00:27:36Lars_GImagine how many you'd get if it said "2 february"
00:27:39 Join einhirn [0] (
00:29:20Bagderwe've got >2000 from the wikipedia article as a comparison
00:30:09linuxstbMaybe that's why we've had a lot of questions about the 160GB today...
00:30:34 Nick B00daW is now known as B--daW (
00:30:57Bagderah right, the article mentions the 160GB classic in the picture text
00:31:41linuxstbI've no idea _what_ it's saying though...
00:31:57Bagder"160 GB is more than enough for lossless music"
00:33:19linuxstbIf only that was true...
00:33:33magmaniacthat's around 500 CDs..
00:33:59Bagdernew pc world article
00:34:51linuxstb"linux-like" appearance?
00:34:55Lars_Gmagmaniac: in flac?
00:34:57hcs"all iPods", yay
00:35:00 Quit Febs (Read error: 110 (Connection timed out))
00:35:26Lars_Ghcs: means lots of people comming to inquiry for the touch, the classic and the fatty :(
00:35:45magmaniacLars_G: yes, roughly
00:35:47*linuxstb stops reading that article at the mention of Furthermore, the Rockbox support forums contain the occasional scary post declaring "Rockbox ruined my iPod!"...
00:36:31Lars_Glinuxstb: shouldn't someone ask for the right to reply that fud?
00:36:51Lars_GI'll test writting to the usb stack in a while :D
00:37:07 Quit morrijr (
00:37:07 Quit lostlogic (
00:37:07 Quit sslashes (
00:37:07 Quit basti (
00:37:07 Quit feisar (
00:37:07 Quit fxb__ (
00:37:20gevaertsLars_G: what device are you using ?
00:37:22Lars_Gdamn. a netsplit
00:37:34NJoinbasti [0] (i=bnc@
00:37:39Lars_Ggevaerts: e260, non R
00:37:47NJoinfxb__ [0] (
00:38:01 Join sander_m [0] (
00:38:16*rasher scolds PCWorld for linking directly to a specific rbutil version
00:38:32Lars_Grasher: they're worse than end users... what you'd expect
00:38:36 Join J [0] (
00:38:51NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
00:39:03NJoinsslashes [0] (i=rmf@
00:39:09NJoinfeisar [0] (
00:39:40sander_mHello. If my rockboxed ipod is mounted on /media/ipod, where do I put the music so the database works correctly? Can I put it anywhere? Does it have to be in /media/ipod/Music? /media/ipod/iPod_Control/Music?
00:40:32Lars_Gsander_m: In short, anywhere will work, but I like to keep it clean
00:40:41Lars_GI use /media/<ipod>/Music for my own
00:40:58midgeygevaerts: svn using fullspeed seems to play nicely with my sansa e200 and mac
00:41:40Lars_Gclear stuff a little for me, full speed is 1.1's 12Mb and high speed is 480Mb right?
00:42:05sander_mThing is, I put a bunch of music under /media/ipod/Music and went to settings>>database and hit "update now". The old music (that I deleted from /media/ipod/iPod_Control/Music) disappeared, but the new songs from /media/ipod/Music wasn't added.
00:42:14Lars_Gmidgey: I hope that's true, my e260's OF is incredibly fidgety on the mac, it sometimes takes me three reboots to have the mac kernel stop panicing and mount it.
00:42:39Lars_Gsander_m: try reiniting the db from scratch, and, did you reboot after the update?
00:42:47gevaertsmidgey: so your earlier problems were probably caused by high speed I guess
00:42:58***Saving seen data "./dancer.seen"
00:43:07Lars_Gfull speed is the default for svn right?
00:43:12 Quit toffe82 (Read error: 104 (Connection reset by peer))
00:43:22midgeyLars_G: i found that the oldest firmware seems to work the best. other versions refused to mount
00:43:36gevaertsLars_G: yes. There are still problems with high speed
00:43:43sander_mLars_G: After installing RockBox or after putting the new music on it?
00:43:45midgeygevaerts: seems likely, although i can't figure out dmesg
00:44:11midgeyLars_G: thats OF version 01.01.11A btw
00:44:39 Join toffe82_ [0] (
00:44:48Lars_Gsander_m: after rebuilding or updating the db-
00:44:48 Nick toffe82_ is now known as toffe82 (
00:46:02sander_mLars_G: No. Do I have to do that after updating the DB?
00:46:18Lars_Gsander_m: As of last version I've seen. yes
00:46:33sander_mLars_G: Okay, thanks.
00:46:59 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
00:47:24gevaertsmidgey: on linux I also get 'uncommon' error messages with high speed, although not while enumerating. I really want to trace what's on the wire, which I can't do until monday, so there probably will be little or no progress on high speed until then.
00:47:37sander_mLars_G: It works. Thanks!
00:47:43 Quit dandin1 ()
00:48:59 Quit DerDome ("Leaving.")
00:49:01 Join Sivart0 [0] (
00:49:14Lars_GBtw I see there's serial mode and drive mode for the rockbox stack right now.
00:49:37Lars_GDo you think it'd be possible in the future to implement both as a multipoint virtual dev? we could control and load rockbox at the same time!
00:50:15Lars_Ggevaerts: I'm naturally wary of high speeds anyhow, my (via) chipset is horribly unstable for ehci, so much I tend to dissable it by hand
00:50:53 Quit JdGordon (Remote closed the connection)
00:51:21gevaertsThey actually can be enabled at the same time right now (uncomment the #define USB_SERIAL in usb_core.h). It currently just echoes back what you send, with case inverted.
00:52:04 Quit ender (" On the contrary, if you never procreate, neither will your kids.")
00:52:09Lars_GVery nice
00:52:47gevaertsThe first real use of that will probably be to send debug output to the host.
00:53:15 Quit robin0800 (" HydraIRC -> <- Organize your IRC")
00:54:01Lars_GYes it'd be lovely for you guys to integrate that with gdb
00:54:33Lars_GBut it opens a host of possibilities.
00:54:42Lars_GIn fact I might try my hand at one I just came up with.
00:55:00 Quit spiorf (Remote closed the connection)
00:55:01gevaertsLars_G: I think that gdb over this usb stack is unlikely to work well.
00:55:03 Quit greatgatsby ("Leaving")
00:55:12 Join JdGordon [0] (
00:55:15 Join FOAD_ [0] (
00:58:51 Quit sander_m ("Leaving")
00:58:51 Quit toffe82 (Read error: 104 (Connection reset by peer))
00:59:26 Join toffe82_ [0] (
00:59:35 Nick toffe82_ is now known as toffe82 (
01:00:25 Quit Arathis ("Bye, bye")
01:00:54 Quit toffe82 (Read error: 104 (Connection reset by peer))
01:01:45 Join toffe82_ [0] (
01:01:46 Quit toffe82_ (Client Quit)
01:02:54 Quit amiconn (Nick collision from services.)
01:03:01 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
01:07:58 Quit gevaerts ("ZZzz..")
01:09:17 Join Hodapp [0] (
01:10:45HodappI just got a current build according to the instructions on the site for a 3rd-generation iPod, and copied firmware and bootloader... but when I reset, I get the Apple logo and then a black screen with the backlight on.
01:10:46 Quit FOAD (Read error: 110 (Connection timed out))
01:10:46 Nick FOAD_ is now known as FOAD (
01:10:59HodappI tried both the automatic install and manual install, with the same results.
01:11:24HodappI can still reset back to factory settings, but I'd like to know what is going on here
01:11:35Hodappdo I need to do something different to reset and start up?
01:12:19*linuxstb wonders why the Player has a font_file in the settings struct
01:13:19JdGordonits just lucky?
01:13:44 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
01:14:04Hodapphm. It seems I can even still turn the Hold switch on at bootup and get normal firmware.
01:17:25linuxstbJdGordon: It's now unlucky again...
01:17:25pixelmaHodapp: I guess you're experiencing a not easy to fix problem on the older Ipods currently (1st to 3rd gen), basically some builds work some won't (freezes or data aborts). Try some of the archived daily builds from the builds page > daily builds > old
01:18:36Hodapppixelma: how old do you recommend going?
01:19:44 Quit w1ll14m|away ()
01:19:56Hodapphm. If I can still get to normal firmware with the hold switch - does that mean that the bootloader is fine, but the Rockbox firmware is not?
01:21:01pixelmathe date is not so important as the problem is older it's just that you can easily chose a bunch of builds to try. But yesterday there was someone with a 2nd gen Ipod who ran into this problem and said the build from the 17th works (not sure if it's always the same for 2nd and 3rd gen though)
01:21:12 Join robin0800 [0] (
01:21:38pixelmayes, bootloader should be fine only the build (which you get as zip) is not
01:23:03 Join DaCapn [0] (
01:23:19 Quit JdGordon (Read error: 104 (Connection reset by peer))
01:24:13Hodappbuild from 2/01 - success
01:25:27linuxstbWhen CabbieV2 is made the default theme, should Rockbox revert to it when the user selects "rockbox_default", or when the user attempts to load an invalid WPS?
01:25:33 Join JdGordon [0] (
01:26:05 Join DerPapst [0] (
01:26:12DerPapstmorning :-)
01:26:41Hodappcan this work as a disk over Firewire? I'm kinda clueless about stuff like this. I've just noticed that when I plug it in with Firewire, it displays a USB icon and then restarts into Disk Mode
01:27:00*DerPapst wonders...
01:27:14DerPapstis the irc-log2html script in svn?
01:27:22DerPapsti can't find it in www
01:28:07Sivart0i think you have to recompile rockbox with USE_ROCKBOX_USB defined
01:28:20Sivart0though i dont know if it would work with firewire =o
01:28:30 Join Febs [0] (
01:28:57linuxstbHodapp: Yes, the 3rd gen has both firewire and usb - Rockbox reboots into disk mode when it detects either.
01:29:13linuxstbSivart0: The new Rockbox USB stack is not for the 3rd gen.
01:29:20 Quit waldo ("Konversation terminated!")
01:29:22HodappOh. So it's my Mac being stupid when it doesn't recognize that a disk is present over firewire?
01:29:30*Hodapp grabs USB cable from random Windows box
01:30:04Sivart0my bad =S
01:30:58 Quit gregzx ("ChatZilla 0.9.81 [Firefox]")
01:31:21linuxstbDerPapst: No, I don't think it is. You'll need to ask Zagor if he wants to share it.
01:31:55DerPapstlinuxstb: aha.. thanks for the confirmation :-)
01:32:15 Quit einhirn (Read error: 104 (Connection reset by peer))
01:34:42midgeyDerPapst: as in the perl code for the irc log reader?
01:38:11midgeycheck the bottom of the irc page. reader.txt should be what you want
01:40:15DerPapstuummm... heh.
01:40:33DerPapstseems lke my glases aren't thick enough yet ^^
01:40:38DerPapstthanks midkay ;-)
01:40:39JdGordonBagder: have a look for libwiimote (or something like that...)
01:41:28Hodapplinuxstb: Any idea why my Mac isn't recognizing the rockbox'd iPod by Firewire?
01:41:54pixelmaDerPapst: check your tab completion while you're at it... :P
01:42:18HodappI wouldn't care ordinarily but it's a bit of a pain to have to charge via Firewire and transfer music via USB
01:43:12DerPapstah well... maybe i should better get some sleep ;-)
01:44:07midgeyat least the first few letters were right
01:46:31pixelmaHodapp: not sure if it helps (don't have an Ipod) but what happens if you first boot into the original firmware and then connect?
01:47:13 Quit JdGordon ("Konversation terminated!")
01:47:21Hodappsame thing
01:47:38*DerPapst thanks midgey and excuses himself by midkay for highlighting :-)
01:48:05midkaynp :)
01:48:23midgeyhappens a fair amount actually
01:48:26*pixelma thinks that "by" was very... german
01:49:37linuxstbHodapp: Did it work before you installed Rockbox?
01:50:16 Join Zarggg [0] (
01:50:21DerPapstpixelma: eh. my grammer is bad.. and i haven't found a nice translation for that ;-)
01:50:45 Quit magmaniac ("Leaving")
01:50:55linuxstbDerPapst: I would say "and apologies to midkay for..."
01:51:05linuxstbI mean "apologises"...
01:51:07 Quit Axio (Remote closed the connection)
01:51:29*DerPapst writes that down into his vocabulary book
01:51:37DerPapstthanks :-)
01:52:14 Join japc [0] (
01:52:14Hodappat this point it appears to be shunning the iPod
01:52:57*Hodapp ponders...
01:53:06*Hodapp tries other cable
01:53:13Sivart0macs can be so cruel ='(
01:53:49HodappYeah. I'd ask ##mac but they'd probably say something like "well you should have put Rockbox on there then LAWL"
01:54:25Sivart0or RTFM <_<
01:54:45Hodapphow's TFM help me?
01:55:05Sivart0no, i mean that the people in mac would tell you that
01:55:13linuxstbHodapp: You could try removing the Rockbox bootloader (run ipodpatcher and select uninstall), and see if your Mac changes its mind.
01:55:39Hodapplinuxstb: But doesn't that kinda make it entirely irrelevant that Rockbox is on the iPod in the first place?
01:56:41pixelmalinuxstb: don't you also have problems with connecting your Ipod to USB on a Mac sometimes which fixes itself with a reboot </wild guess>
01:57:12 Quit midgey ()
01:57:28Sivart0brb, booting into linux >_>
01:57:30 Quit Sivart0 ("ChatZilla 0.9.81 [Firefox]")
02:01:54linuxstbpixelma: That was (I think) related to the Rockbox USB detection code breaking the Mac's USB stack sometimes. That detection code isn't used on the 3g.
02:02:16 Join travis_ [0] (
02:02:24linuxstbHodapp: I'm not sure what you mean...
02:02:34 Nick travis_ is now known as Sivart0 (
02:02:48Sivart0back :D
02:03:04Hodapplinuxstb: if I remove the Rockbox bootloader, how is that different from just resetting the iPod? I can't get into Rockbox without the bootloader, can I?
02:03:27 Quit DaCapn (Read error: 110 (Connection timed out))
02:04:47linuxstbHodapp: If your ipod still doesn't connect, even with Rockbox removed, then that would seem to suggest Rockbox isn't to blame. Although as pixelma suggested, maybe a reboot of your Mac might be useful, just to reset everything to a known good state.
02:08:38*Hodapp rebooting...
02:09:55 Quit BitTorment ("Do directly to guantanamo bay. Do not pass go. Do not collect your human rights.")
02:17:10Hodappbehold, a reboot of the Mac fixed it
02:17:53Sivart0i uninstalled rockbox from my ipod just now ><
02:18:03Sivart0i gotta stop messing with stuff >_>
02:19:22 Quit ol_schoola ()
02:20:19Hodappyay, it works in disk mode too
02:23:50 Quit Jon-Kha (Remote closed the connection)
02:24:30 Quit robin0800 (Read error: 104 (Connection reset by peer))
02:26:51 Quit DerPapst (Read error: 110 (Connection timed out))
02:38:15 Quit johnh51 (No route to host)
02:38:36 Quit linuxstb (Read error: 110 (Connection timed out))
02:40:31 Join corevette [0] (
02:43:02***Saving seen data "./dancer.seen"
02:47:04 Join barrywardell [0] (n=barrywar@
02:47:31 Quit x1jmp (Remote closed the connection)
02:47:44 Quit Sivart0 ("ChatZilla 0.9.81 [Firefox]")
02:49:39 Join midgey [0] (
02:53:49 Quit hannesd (Read error: 113 (No route to host))
02:58:42 Quit XavierGr (Nick collision from services.)
02:58:53 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
03:04:05 Quit amiconn (Nick collision from services.)
03:04:11 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
03:08:31 Quit Nico_P (Remote closed the connection)
03:11:13 Quit dan_a (Read error: 110 (Connection timed out))
03:17:38 Join johnh51 [0] (
03:20:20 Quit fasmaie (Read error: 113 (No route to host))
03:29:07 Part pixelma
03:42:04 Quit johnh51 (Read error: 113 (No route to host))
03:43:34 Join johnh51 [0] (
03:43:45 Join S [0] (
03:47:54 Join crzyboyster [0] (n=4b596e0c@gateway/web/cgi-irc/
03:48:46 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
03:50:44crzyboysterWill vmware work on a computer with 256 mb ram? And what other requirements does it have because my vmware says that there is not enough "physical memory"...
03:51:07saratogai'd guess it won't work well
03:51:33saratogamight be ok for the command line though
03:53:56B--daWi want shit on my nano 2.
03:53:59 Nick B--daW is now known as B00daW (
03:54:22B00daWyou guys are all capable of doing that in the next 5 mins, right?
03:54:27B00daWsuch smart guys.
03:55:00crzyboysterHow do I launch the command line version? And mine refuses to launch. It freezes saying the following: "Not enough physical memory is available to power on this virtual machine."
03:56:23saratogacrzyboyster: I'm not sure how the version we distribute works
03:56:34saratogait may default to using x-win
03:56:41saratogain which case you will need a fair amount of ram
03:56:47 Quit conando (Read error: 113 (No route to host))
03:57:20crzyboysterWho handles the rockbox vmware distribution?
03:57:36saratogai have no idea
03:57:41saratogamaybe check the wiki and see who edited it
04:01:24 Quit crzyboyster ("CGI:IRC (EOF)")
04:03:35saratogaanyone around with a grayscale ipod?
04:03:44Hodappsaratoga: I think so?
04:03:59saratogacould you try a build for me and tell me if the menus look right
04:04:36Hodappno :\ I had to program it on a Windows box because I couldn't make it work right on OS X and I'm on my Macbook now
04:05:19saratogai can provide you with a download link
04:05:29saratogaif you tell me what device you're running
04:05:49Hodapp3rd gen iPod
04:08:15saratogayou must rename your current .rockbox folder
04:08:35saratogajust rename it, try this one, and when you're done, you can delete it and rename your original folder back
04:13:24saratogaHodapp: any results?
04:14:28HodappI can't test it right now... read a few lines back
04:15:21saratogaHodapp: you might have mentioned this before i compiled you the build . . .
04:15:49Hodapp22:01 < saratoga> could you try a build for me and tell me if the menus look right
04:15:52Hodapp22:01 < Hodapp> no :\ I had to program it on a Windows box because I couldn't make it work right on OS X and
04:15:55Hodapp I'm on my Macbook now
04:16:17saratogawhich is why i offered to compile it for you
04:16:23saratogaso that you could upload it with your macbook
04:17:01Hodappsorry, what I meant was that I couldn't program from my Macbook
04:17:22saratogado you mean update rockbox?
04:17:28saratogaif so, why not?
04:17:37saratogaa macbook will work fine
04:17:50saratogaassuming it was made in the last 8 years anyway
04:17:56HodappIn theory it will work fine.
04:19:00Hodappbut for me it had issues and I couldn't get the access I needed to the iPod in disk mode, and the necessary software could not communicate properly
04:19:31Hodappand it was looking like I needed to use a Windows box to format it to FAT32 anyway so I figured I may as well update from there as well
04:22:08*Hodapp shrugs
04:22:14HodappI have a Linux box at home I can use
04:23:51saratogaso you couldn't even mount your ipod in macos?
04:23:54saratogahow do you add music
04:24:40*Hodapp shrugs.
04:26:00HodappI suppose my original issue would pertain more to getting the bootloader on there than to getting firmware on there.
04:26:36Hodappnow, I do know this: I tried the most recent current build and it didn't work; a build from the 1st of the month worked just fine
04:26:57Hodappis your build based on the most recent?
04:29:14saratogai'm about to commit it
04:29:24saratogaso it'll be most recent in about an hour
04:29:56HodappI have a sneaking sensation that this build won't work if the build preceding it didn't work
04:30:09saratogaare the 3g builds broken lately?
04:30:12saratogai hadn't heard that
04:30:16saratogathough it wouldn't surpise me
04:32:55HodappI just got this iPod like 5 hours ago and tried Rockbox, so I dunno
04:38:42 Quit miepchen^schlaf (Connection timed out)
04:38:54 Join miepchen^schlaf [0] (
04:41:19 Join webguest77 [0] (n=4c10ea95@gateway/web/cgi-irc/
04:42:17webguest77hi i have a question can you guys please give me a link or tell me where i can get the rockbox "spanish" voice file?
04:43:01 Join BHSPitLappy [0] (n=Steve-O@unaffiliated/bhspitmonkey)
04:43:07***Saving seen data "./dancer.seen"
04:43:14 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
04:43:51 Quit Thundercloud (Remote closed the connection)
04:44:45webguest77please help if you guys know?
04:50:38 Join RoC_MasterMind [0] (
04:51:15 Quit webguest77 ("CGI:IRC (EOF)")
04:58:47saratogadefault theme commited
05:01:45 Join ol_schoola [0] (
05:12:40saratogahmm it would seem the mrobe 500 exists
05:42:33 Quit nicktastic (Read error: 110 (Connection timed out))
05:42:56 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
05:43:20 Quit jcollie (Read error: 104 (Connection reset by peer))
05:43:56 Quit saratoga ("CGI:IRC (EOF)")
05:46:26 Join Vikas [0] (n=chatzill@
05:46:53Vikaswhen will be classic supported?
05:48:03 Quit Vikas (Client Quit)
05:55:58 Nick B00daW is now known as B--daW (
06:01:15 Join jcollie [0] (
06:02:13 Quit barrywardell ()
06:04:53 Join fdtoonsman [0] (n=4c710dd1@gateway/web/cgi-irc/
06:05:01fdtoonsmanhello people!
06:05:45fdtoonsmanhello people?
06:06:15advcomp2019hello fdtoonsman
06:06:35fdtoonsmanI installed rockbox a week ago, and i have a few problems.
06:06:54fdtoonsmancan anyone help?
06:07:15advcomp2019fdtoonsman, we are not into reading minds
06:07:58advcomp2019plus have you tried the latest build
06:08:05fdtoonsmanwell, first of all, I need to know if it's possible to play mp4 files like on the regular iPod
06:08:15fdtoonsmanand yes, i update my build almost every day
06:09:18advcomp2019fdtoonsman, no.. you only can play mpeg and mpeg2 videos in rockbox
06:09:50fdtoonsmanaw, why not? they should totally make an MP4 player, because that's what the ipod uses, and converting is a hassle!
06:10:12fdtoonsmanit totally makes sense... is it being worked on?
06:12:16fdtoonsmanalso, my converter is being strange.
06:17:17 Quit fdtoonsman ("CGI:IRC")
06:29:00 Quit nicktastic (Read error: 110 (Connection timed out))
06:34:04 Join Doomed_ [0] (
06:34:53 Quit alienbiker99 (Read error: 104 (Connection reset by peer))
06:35:38 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
06:38:39 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
06:43:05 Nick Doomed_ is now known as alienbiker99 (
06:43:08***Saving seen data "./dancer.seen"
06:49:55 Join ackbahr [0] (
07:07:13 Quit midgey ()
07:10:46 Join midgey [0] (
07:12:53*flyback ya know if something were to happen to me, i'd have to say one of the best things in my life was discovering irc and making friends with all of you
07:13:23zegeez i hope thats sarcastic or something :p
07:13:55flybackwell that was a global for all my channels :P
07:14:03zestill.. heh
07:24:53 Quit nicktastic (Read error: 104 (Connection reset by peer))
07:25:12 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
07:30:39*flyback goodnight all
07:32:20 Quit ackbahr (Connection reset by peer)
07:37:34 Join nicktastique [0] (n=nick@unaffiliated/nicktastic)
07:49:48 Quit nicktastic (Read error: 110 (Connection timed out))
07:54:27 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
07:55:01 Quit fyrestorm (Read error: 113 (No route to host))
07:56:07 Quit midgey ()
07:59:35 Join fyrestorm [0] (
08:01:32 Quit EspeonEefi ("さよăȘら")
08:05:50 Quit saratoga ("CGI:IRC")
08:07:23 Join ch4os_ [0] (n=ch4os@unaffiliated/ch4os/x-059673)
08:22:26fyrestormis it possible to make rockbox read apev2 tags on SPC files?
08:22:29 Quit nicktastique (Read error: 110 (Connection timed out))
08:23:31 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
08:26:57 Join EspeonEefi [0] (i=espeonee@CERULEANCITY.MIT.EDU)
08:27:58 Quit fyrestorm (Read error: 113 (No route to host))
08:30:37 Join spiorf [0] (
08:31:37 Nick Neskaya is now known as Uh-oh (i=featherd@wiktionary/neskaya)
08:31:52 Join fyrestorm [0] (
08:32:37fyrestormplugging in ipod with current build is making my system restart
08:33:31amiconnbah, crappyv2
08:33:59fyrestormive gotten two bsod's by just plugging it in
08:35:21Horschtfyrestorm, boot to the OF first, then plug it in
08:35:23fyrestormtrying yesterday's build
08:36:09 Join Rob2222 [0] (
08:36:19fyrestormok, in OF, plugging it in
08:36:30fyrestormim still alive.
08:36:57fyrestorm(also, im using ipod vid 5th 30gb)
08:39:08fyrestormok, using yesterday's build. plugging it in
08:39:28fyrestormstill alive :)
08:40:16fyrestorminstalling recent build
08:41:02fyrestormk, restarting it and will plug it in in a second
08:41:04 Quit midkay ("Leaving")
08:41:22fyrestormdoing it now
08:42:32 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
08:43:11***Saving seen data "./dancer.seen"
08:43:34 Quit fyrestorm (Read error: 113 (No route to host))
08:45:23amiconnBah, now my mini switched to that crappyv2 theme, and there is *no line selector* anymore, so I can't set it back :(
08:46:55amiconnIt's a bug in the line selector assignment.
08:46:56 Join fyrestorm [0] (
08:47:22amiconnThe theme selection assigns a non-existing line selector type for greyscale targets
08:47:34*amiconn thought saratoga tested this stuff :(
08:48:25amiconnI also wonder why it swoitched to crappyv2 at all, as the commit says it should not change existing settings :(
08:48:52 Join mf0102 [0] (n=michi@
08:50:56amiconnAnd I can't get rid of the backdrop, grrrr!!
08:53:04 Quit Horscht ("IRC is just multiplayer notepad")
08:53:15 Quit Rob222241 (Read error: 110 (Connection timed out))
08:57:25amiconnSettings->Clear backdrop, and the backdrop is gone. But at next boot it is back, even though config.cfg doesn't set it??!
08:58:26pixelmaI thought it should be set in config.cfg then because it differs from the default...
08:59:36 Quit nicktastic (Remote closed the connection)
09:00:38 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
09:01:37amiconnSame crap on H10. Rockbox is now unusable for me on colour targets :(
09:02:00amiconnBlack text on (nearly) black backdrop
09:02:11 Join Evilnick_ [0] (
09:05:35 Quit moos (Read error: 110 (Connection timed out))
09:11:08 Join Jon-Kha [0] (
09:12:00pixelmaI just tested in a Video sim where I had set iCatcher before as theme -> it did change to cabbiev2 although it shouldn't because I hadn't used the default. In this case colours changed correctly (line selector and foreground colour) but the font stayed, i.e. it was set to nimbus-19 which iCatcher uses
09:15:37pixelmahmm... and changing back doesn't reset colours. The generated iCatcher.cfg does not contain the "foreground colour", "background colour" lines which I think it did before (maybe this is due to the other related commit of yesterday?)
09:19:23pixelmathis is actually good in the menu here because the background isn't cleared either and so I don't end up with light grey on white background which I do in the WPS then though
09:22:38 Quit Evilnick (Read error: 110 (Connection timed out))
09:22:40pixelmaI can see "Foreground Color: 000000" and "Background Color: FFFFFF" for iCatcher in WPSLIST btw. so somehow it is skipped now while generating the cfg
09:25:28 Join n1s [0] (
09:32:34*amiconn summons saratoga
09:32:45 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
09:36:01 Join tvelocity [0] (
09:37:06 Quit jhulst ("Konversation terminated!")
09:39:20 Join petur [50] (n=petur@rockbox/developer/petur)
09:42:52 Join Axio [0] (
09:42:56 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:46:51pixelmaI'd almost vote for a revert until those IMO important bugs are sorted out... (on clolour targets it seems too easy to get a (almost) unreadable colour combination especially since it sets cabbiev2 even if you had set it to something else than the default before, the backdrop issue aso...)
09:48:07pixelma*colour ...
09:49:46*linuxstb thinks we need a "wait 48 hours before committing to give people chance to comment" rule when people post patches saying "is this OK to commit?"
09:50:08 Join advcomp2019_ [0] (n=advcomp2@
09:50:18 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
09:50:21pixelmayeah, it really made me wonder
09:50:38linuxstbI had a lot of comments I was going to make, but didn't have time last night to do it...
09:51:47 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
09:52:24 Join moos [0] (
09:54:19pixelmaah, I understand the iCatcher -> cabbiev2 issue better now. It doesn't override the setting of your cfg as I thought first but iCatcher doesn't set a background (the new default will be used) and not the colours anymore (new default) and no icons - only the WPS would have changed
09:54:59pixelmaso the menu looked like cabbiev2 with iCatcher font
09:55:23amiconnI think there are now just 2 bugs left. (1) The wps packager somehow leaves out the colours even on colour targets
09:55:47amiconn(2) The settings code doesn't save the setting for "no backdrop" even though this is no longer the default
09:55:57amiconnThis is actually a settings code bug
09:56:14pixelmaand what about the iconset?
09:56:42linuxstbShouldn't the other themes include "default iconset" in their cfg now?
10:00:31pixelmafirst "bug report" thread about it in the forums... "Help! No Cursor!" on a 4th gen Ipod greyscale...
10:02:19linuxstbIs that the bug amiconn fixed?
10:03:13pixelmayes, I'll answer
10:05:46 Join midkay [0] (n=midkay@rockbox/developer/midkay)
10:08:34linuxstbamiconn: Are you looking at any of those bugs you mentioned? If not, I have a little time now.
10:10:48amiconnI am trying to fix (2)
10:11:25amiconnThe settings code is kinda strange - and there already seems to be a hack-around for the iconset
10:11:26linuxstbOK, I'm looking at 1...
10:11:40amiconn(by setting the filename to "-.bmp" instead of empty string)
10:15:26 Join stel [0] (n=steve_la@
10:15:59amiconnWhere are those theme .cfg files generated?
10:16:50pixelmalinuxstb: about (1) - probably related to saratoga's earlier commit yesterday to
10:17:22pixelmathey are created by using WPSLIST
10:18:20linuxstbpixelma: Yes, found, fixed and committed ;)
10:19:22pixelmawhy is there one thing named "filetylecolor" btw... :)
10:20:01linuxstbThat's very odd - a consistent typo...
10:20:45amiconnc&p typo
10:21:42linuxstbWhat do people think the "default_rockbox" theme .cfg file should do? Seems odd to have one theme called "CabbieV2" which is the real default, and another theme called "rockbox_default" which isn't...
10:23:01 Join chuff [0] (
10:23:14chuffhow do i rockbox my ipod shuffle
10:23:32linuxstbchuff: You can't - Rockbox doesn't work on the shuffles.
10:23:59linuxstbThe front page of the Rockbox website lists all the supported players.
10:24:02chuffthe point of rockbox is so that you don't have to use itunes software... right?
10:24:07amiconnlinuxstb: Does the wpslist code allow to output an empty string for a setting?
10:24:32linuxstbamiconn: I don't know, let me test...
10:24:39 Join FOAD_ [0] (
10:24:44amiconnRight now it e.g. sets an iconset to "-" which translates to /.rockbox/icons/-.bmp
10:25:10amiconnVery odd if someone actually puts a file named "-.bmp" there (that's a valid filename)
10:26:19linuxstbYes, we should fix that. If I do "iconset:" in WPSLIST, then the wpsbuild script complains when trying to copy that non-existent bmp file...
10:27:16*linuxstb notices a whole load of TABS in
10:28:55 Join desowin [0] (n=desowin@atheme/developer/desowin)
10:29:03amiconnThe settings code does handle filename settings in an odd way
10:29:38amiconnThere are 2 possible types, filename settings with prefix and suffix, and such without both (used for recording dir and playlist catalog dir)
10:29:58amiconnThe ones with prefix and suffix are decomposed on load, and recomposed on save
10:31:07chuffhow do i move songs from computer to ipod?
10:31:19chuffitunes is being very difficult
10:31:50linuxstbchuff: This channel is for discussion of Rockbox, not general ipod discussion.
10:32:04*pixelma looks around whether other themes would need a foreground/background colour definition in WPSLIST now that black on light blue isn't default anymore...
10:33:26linuxstbpixelma: I guess there are a lot of other settings that themes must now include - icons, viewer icons, selector bar, ...
10:34:10pixelmahmm, right
10:35:05linuxstbamiconn: There is a change needed for to accept empty strings - the code is using "if ($iconset)", which should be "if (defined($iconset))"
10:36:12linuxstbOr maybe we should make wpslist output empty strings for all theme setting variables, even if they're not defined in WPSLIST...
10:36:49amiconnWell, not all theme settings are valid on all targets
10:36:57amiconn(like colours on mono targets)
10:38:53amiconnI wonder why saratoga removed colours for greyscale targets, btw
10:39:42amiconnThe current UI doesn't allow setting the greylevels, but the low level code is there, and imho we should offer those settings
10:40:42amiconnEven the solid "colour" line selector would be possible on greyscale targets. Gradient obvously won't work with just 2 bits
10:40:44pixelmalinuxstb: so... should I wait a bit? It's just that I thought this could be a simple enough thing even I could do to help fixing...
10:40:45 Quit FOAD (Read error: 110 (Connection timed out))
10:40:45 Nick FOAD_ is now known as FOAD (
10:43:13***Saving seen data "./dancer.seen"
10:43:17linuxstbamiconn: I'm sure saratoga just followed what was in settings.h - where the colours are protected by HAVE_LCD_COLOR. But I agree it would make sense to offer them.
10:44:00 Quit ompaul (Remote closed the connection)
10:44:06linuxstbpixelma: I'm not sure... I think we're all (you, amiconn and me) are looking at related things.
10:44:21pixelmathe lang files would even contain foreground/background colour on greyscale targets, these are just not there on mono LCDs
10:45:49 Join dan_a [0] (n=user@
10:46:17 Join stripwax [0] (n=Miranda@
10:47:00linuxstbamiconn: Should I commit a change to to enable empty strings - e.g. "iconset: ", "backdrop: " to be output to the theme .cfg ?
10:47:21amiconnI think so
10:48:03amiconnI'm still not 100% sure about that settings code. What does strncasecmp() return if one of the strings is empty? Just 0?
10:50:09linuxstbI would guess +/-1 - i.e. "" is < "a"
10:52:14amiconnYeah, looks like it does
10:53:02 Join qwedsa_ [0] (
10:56:45linuxstbSeems that doesn't copy the "filetype colours:" file if it's defined (which it currently isn't for any theme)
11:04:33amiconnOkay, saving empty filename settings works now, and produces what one would expect in config.cfg, e.g. "backdrop:"
11:06:05pixelmathe thing is that not all included themes might *need* to set some things like selector type and so on, or should I just assume all do (to have them look like before)?
11:07:41pixelmaand for no backdrop it should read "Backdrop: none" in WPSLIST then?
11:07:54linuxstbI thought it should be "Backdrop:"
11:08:13linuxstb(but I could be wrong...)
11:09:10linuxstbamiconn: Would a trailing whitespace - i.e. "backdrop: " be OK in a .cfg?
11:10:26 Quit qwedsa (Read error: 110 (Connection timed out))
11:11:04amiconnThat's no problem
11:12:18amiconnEven multiple spaces are okay, as the parser skips spaces between : and the value, but only plain (0x20) spaces are allowed
11:13:01linuxstbI think I'll fix so it doesn't output it though (and so it doesn't need one in WPSLIST)...
11:15:21 Join hannesd [0] (
11:22:12 Join devo132 [0] (n=4582a153@gateway/web/cgi-irc/
11:22:28linuxstbI've just committed my changes to, and adding empty settings for iCatcher. The other themes still need fixing.
11:26:40amiconnlinuxstb: The default viewer iconset must not be empty, otherwise you won't get any viewer icon
11:27:13 Join nplus [0] (
11:27:18amiconn(i.e. viewer icons are always loaded from disk)
11:28:28pixelmalinuxstb: ok, something for me then
11:28:29amiconnExcept on player - that still needs fixing
11:28:55amiconn(player shows *no* icons for viewer-supported files right now)
11:30:09 Quit devo132 ("CGI:IRC")
11:31:26linuxstbamiconn: Hmm, it works for me - when I select iCatcher, I get the default icons.
11:31:31linuxstbAh, viewers...
11:31:56amiconnCheck the icons for .jpg, .mpg files etc...
11:32:21linuxstbSo what should icatcher set in WPSLIST if it wants the default icons?
11:33:12 Join waldo [0] (
11:34:11linuxstbI'm guessing "/.rockbox/icons/viewers.bmp" ?
11:34:57linuxstbOK, I'll fix.
11:35:36 Part dan_a
11:36:23 Join webguest48 [0] (n=4582a153@gateway/web/cgi-irc/
11:37:27linuxstbSo I guess I should also revert the change that allows empty values for "viewers iconset" in ?
11:39:13 Quit webguest48 (Client Quit)
11:41:24 Join devo1324 [0] (n=4582a153@gateway/web/cgi-irc/
11:41:38BigBambiIs the old default still called rockbox_default? If so, how about a rename to something like "Rockbox Classic"?
11:42:26devo1324 hey can i have write permission for the wiki i've been dying to fix some errors
11:42:45BigBambiWiki name?
11:42:51 Quit GodEater (Read error: 104 (Connection reset by peer))
11:43:36BigBambiJust like that?
11:43:42BigBambiOr with capitals?
11:43:55devo1324let me check
11:44:40devo1324for the d and g
11:44:55BigBambiCould you jusdt type your actual username for me please
11:44:56pixelmaBigBambi: you could also check the "recent wiki activity" ;)
11:45:17BigBambipixelma: I know, but I think it important people realise accuracy is important :)
11:45:48BigBambidevo1324: OK, done. No spamming now!
11:46:15devo1324thank you1
11:47:01linuxstbdevo1324: What errors have you found?
11:47:30pixelmaabout the included WPSs... there are some, probably old, ones for the 160x128 greyscale screens. First they are named inconsistently (should be 160x128x2) and second - those could easily have a colour version with the classic colours set. What do people think?
11:49:10*amiconn isn't sure what to do about the viewer icons on the player
11:49:25amiconnThe original idea was to add a minimalistic bmp loader
11:50:03linuxstbpixelma: Maybe we should remove some of the older ones... I don't know what others think, but I would like a small (but varied, and not necessarily the same across targets) set of themes in SVN.
11:50:23amiconnBut (1) that doesn't allow falling back to hardware characters if an icon isn't displayable, and (2) it's not like we'll see dozens of new file formats supported on player
11:50:40 Join Lear [0] (i=chatzill@rockbox/developer/lear)
11:51:03pixelmalinuxstb: I'll leave that for later then
11:51:21amiconnSo maybe we should just map bmp index -> built-in icon?
11:52:06 Join barrywardell [0] (n=barrywar@
11:52:11 Join crazybus [0] (
11:52:26linuxstbpixelma: Yes, I think the most important thing is to make the bundled themes actually work...
11:55:51 Join Domonoky [0] (n=Domonoky@
11:55:55 Quit devo1324 ("CGI:IRC (Ping timeout)")
11:56:12pixelmaindeed, I'll copy things from your iCatcher example then and include all settings that might be necessary for a certain theme (e.g. line selector type for greyscale themes to) and with the classic colours where none are set yet - to be on the safe side and to make them look like before
11:56:53pixelmadoes someone know the hex value of rockbox blue off the top of their head?
11:57:54BigBambiThat would be excellent random knowledge :)
11:58:30amiconnLCD_DEFAULT_BG, should be in lcd.h
11:58:48crazybuscan you edit the manual without installing tex and subversion and all their deps (I have to use a different computer due to my main one being broken and I can't install on this one)
11:59:14BigBambicrazybus: Yes, but you can't then check it still compiles or check the output
11:59:34crazybusthanks BigBambi
12:00:15crazybusis the main page for themes now that you can't upload to ?
12:00:33BigBambicrazybus: That always has been the main rockbox page has in the past been entirely unofficial
12:01:00BigBambiWhen it relaunches it will be official
12:01:48linuxstbpixelma: icatcher also needs a default selector...
12:02:44pixelmaprobably "bar (inverse)" it is then
12:03:43crazybusdoes anyone know if there will ever be a option in the spectrum emulator to have letterbox bars or rotate the screen 90 degrees. The current scaling method words hard to read on a sansae200 screen
12:04:02BigBambiIf someone does it...
12:04:04linuxstbcrazybus: No-one has touched the spectrum emulator for months...
12:06:03crazybusthanks again everyone. Another question. Can you change what order items are on the main menu without recompiling. I would like to put resume playback at the top so I resume something without looking at the screen
12:06:08linuxstbIs there a wiki page that describes how to package a theme zip?
12:06:51 Join eigma [0] (n=cat@
12:06:51BigBambicrazybus: no
12:06:58*linuxstb finds ThemeGuidelines
12:07:38crazybusspeaking of the wiki. I have an account. What do I do to activate it to edit
12:07:59linuxstbYou ask in here for it to be activated...
12:08:03BigBambiTell me your wiki name
12:09:06 Nick crazybus is now known as crazy_bus (
12:09:41 Join seablue [0] (
12:10:07BigBambicrazy_bus: OK, done
12:10:13crazy_busthanks BigBambi
12:11:28pixelmathe old rockbox_default that's also listed in WPSLIST also needs these settings right?
12:12:05 Join gregzx [0] (
12:13:09 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
12:13:59crazy_busit says that rockbox has replaygain. But I can't seem to find the option for it in the manual or sound settings. Does it exist?
12:15:33crazy_busor is it just enabled by default
12:16:04BigBambiyes, it exists
12:16:14BigBambiI think the option is in playback or some such
12:18:00crazy_busthanks. I see it now
12:18:46pixelmalinuxstb: do you think the <rwps>s also need the things set? Currently there is no way to use e.g. a different font there but they already define one ("just in case"?)
12:18:53linuxstbpixelma: That's what I'm not sure about - CabbieV2 is now the default...
12:20:01pixelmaI'd think the classic, text based on should be included but maybe as BigBambi suggested it could be renamed
12:20:48linuxstbI can't see a reason to include it, apart from as a hard-coded default when all else fails.
12:20:50BigBambiI think it should stay, but being called rockbox_default now is hella confusing
12:21:15BigBambilinuxstb: Well, some people like it...
12:21:25BigBambiI wouldn't care personally if it were taken out and shot
12:21:40linuxstbBigBambi: Then we should ship it as a real wps, rather than as a virtual file which selects the hard-coded default.
12:22:11pixelmait's really not that bad on my c200 for example...
12:22:35pixelmathough I admit that I use my own :)
12:22:36linuxstbI agree, it's fine on small LCDs
12:23:03 Join d34df00d [0] (n=d34df00d@
12:23:18 Part d34df00d ("Konversation terminated!")
12:24:30pixelmabtw. wasn't there an unanswered question earlier about which WPS the parser should fall back now if there are syntax errors? Actually I don't like that fallback in general
12:24:48 Quit axionix (Read error: 113 (No route to host))
12:38:32 Join qwedsa [0] (
12:39:58stripwaxSo I thought I'd try enabling USE_ROCKBOX_USB on my 5g build. Windows detected the USB Mass Storage device when I connect it (great), but then seems to have hung trying to assign a drive letter. Is that expected right now?
12:40:29stripwaxAh, windows just came back to life - but still no drive letter :|
12:40:35 Join tessarakt [0] (
12:41:23amiconnDid you enable highspeed as well? Because, that doesn't work and causes the symptoms you just described
12:41:44amiconn(explorer appearing to hang for several minutes, non-working access)
12:42:04stripwaxamiconn - not intentionally - do I need to set something to DISable highspeed?
12:42:55LearI briefly tried highspeed on an e200 yesterday. Connect and read worked fine. Didn't dare trying anything more...
12:43:04amiconnJust make sure have have latest svn, especially not r16366 or r16367
12:43:15***Saving seen data "./dancer.seen"
12:43:25stripwaxHm, I disconnected and after rockbox did the Scanning Disk, I got Data Abort at 000089A4 ..
12:43:50stripwaxyep, latest svn. or, at least, let me check that I don't have some crazy patch in my build anywhere.
12:44:20 Part stel
12:45:57 Join stel [0] (n=steve_la@
12:47:33crazy_busI remember there was a alarm patch in custom builds that was dropped because a alarm had being added to offical builds. Was I mistaken as I can't seem to find that functionality anywhere in a default build
12:47:34stripwaxwill try a clean rebuild - the only patches I have are the pacbox on cop and 5g 60gb battery calibration
12:48:18stripwaxspeaking of which - does anyone want to commit the pacbox on cop patch or should I do some more work there before it's committable?
12:49:24pixelmacrazy_bus: whether there is an alarm setting or not is also dependent on your player's capabilities
12:52:32stripwax(like emulating pacman audio, say)
12:53:34crazy_busthanks pixelma.
12:54:22crazy_busstripwax: not that I can commit it but what is packbox on cop, is cop short for some type of music player?
12:55:06pixelmawhat player do you have? And do you remember the flyspray entry you are talking about (maybe search for closed tasks as well)?
12:55:34 Part stel
12:56:08crazy_buspixelma: me? I found the patch but it seems it went out of sinc. And the other alarm patch was only for players with a wakeup feature. Which I don't have; sansa e200
12:56:10 Quit qwedsa_ (Read error: 110 (Connection timed out))
12:57:37 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:58:40 Join stel [0] (n=steve_la@
12:59:15 Part stel
13:00:14 Join Mathiasdm [0] (
13:01:22LearHm, made a logf build, but while there is a logf entry in the debug menu, it doesn't show anything. Nor is there any output on the remote...
13:01:26LearA known problem, perhaps?
13:02:31amiconnIt's desired behaviour for quite a while
13:02:55amiconnYou need to enable logf output in the source file(s) you want to debug
13:03:39amiconnThis was introduced to avoid unrelated clutter in logf output
13:07:47LearTypical that the problem I'd like to look at doesn't seem to appear on my target with an LCD remote...
13:08:06 Join Thundercloud [0] (
13:11:36 Quit ompaul (Read error: 104 (Connection reset by peer))
13:12:26 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
13:15:27 Join Mathiasdm2 [0] (
13:16:22 Join Mathias [0] (
13:17:28 Join MethoS- [0] (
13:17:37 Quit Mathiasdm (Nick collision from services.)
13:17:42 Nick Mathias is now known as Mathiasdm (
13:19:00stripwaxcrazy_bus - that patch puts the pacman emulation engine onto the cop (coprocessor cpu) if available, which frees main cpu for codec processing so you can still listeb
13:19:13stripwaxlisten, to music, while playing hte game
13:20:20 Nick S is now known as conando (
13:22:22amiconnlinuxstb: It seems that the viewer icons default is hardcoded, but then I wonder why it doesn't seem to work...
13:24:41 Quit XavierGr (Nick collision from services.)
13:24:54 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
13:25:11pixelmahmm... another thing. Some themes don't have an rwps and if you chose one of those now it'll use the cabbiev2.rwps which can look odd if you're using a different font (especially on the Iaudios where cabbiev2 uses a backdrop in the WPS)...
13:26:23amiconnThis gui code is a maze :\
13:28:23pixelmashould I just set the rwps of all others to the old default, might be the better option?
13:28:57Bagderpixelma: sounds like a decent thing to do
13:30:55pixelmaok, all others is only "Rockboxed" here... :D
13:32:51 Join BitTorment [0] (
13:34:47pixelmawhich actually might work together with the cabbiev2.rwps except on the H300
13:34:51 Quit Mathiasdm2 (Read error: 110 (Connection timed out))
13:38:35amiconnThe 'Open with...' submenu doesn't scroll its items...
13:40:24 Quit MethoS- (Read error: 110 (Connection timed out))
13:45:17 Join DerPapst [0] (
13:45:24crazy_busis it just my player (sansa e200) or does the gameboy and spectrum plugins both play sounds with lots of crackling added in
13:50:24 Join MajorC [0] (
14:00:00 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
14:04:05 Join Honkboy [0] (
14:06:47 Part Honkboy ("Leaving.")
14:07:09*amiconn summons JdGordon
14:10:08*gevaerts thinks that the main problem with cabbiev2 on c200 is that the LCD is not good enough
14:11:35 Join gregzx_ [0] (
14:11:44 Quit gregzx_ (Client Quit)
14:13:34pixelmamy main grep with it is the font (especially on the c200) on others as well
14:14:58gevaertsMost letters are OK, but M and w are
14:16:11gevaertsThe LCD problem I see is that for most of the brightness and contrast range the backlight shines through the black background
14:16:20pixelmaI don't like that the font is so lightweight, it is one point taller that the one I use (nimbus-10) and it's less readable
14:16:50 Join Monga [0] (
14:17:05amiconngevaerts: What do you expect from a CSTN panel?
14:17:43amiconnBesides, one thing I don't like with cabbiev2 is that it's bright text on black (negative display)
14:19:23gevaertsamiconn: not much. I know I got the cheapest player available. It just shows that a dark background doesn't work well on all players
14:19:45rasherAre there any good, less skinny fonts at the same height?
14:19:45 Quit Axio ()
14:20:23*gevaerts just switched back to rockbox_default
14:21:04pixelmarasher: yes, nimbus-11 ;)
14:21:17rasherThat settles that, then
14:21:23rasherNimbus is lovely
14:22:37*gevaerts notices that rasher doesn't use a cron job to update his ircstats page
14:23:41pixelmarasher: the original creator and the most lively participant(s) from the forums said "nimbus is not an option" though I can't remember them giving a reason
14:24:36rasherDoes the current font have better glyph coverage than nimbus?
14:27:43rasherI suppose I should be able to answer that
14:28:57pixelmanot sure, nimbus is already quite ok (having greek, cyrillic and hebrew support), that I think the sazanami-mincho one they chose for the c200 is slightly better there
14:29:17 Quit gregzx (Read error: 110 (Connection timed out))
14:34:20 Quit eigma (Read error: 110 (Connection timed out))
14:37:09 Part pixelma
14:43:20***Saving seen data "./dancer.seen"
14:46:39rasherSazanami-mincho includes a bunch of CJK
14:47:22rasherAs well as katakana and hiragana
14:47:49rasherNo hebrew though
14:47:59rasherWhich nimbus on the other hand has
14:51:51 Part Monga
14:52:19 Quit crazy_bus ("ChatZilla [Firefox]")
14:55:45 Join Horschti [0] (
14:56:31 Quit Horscht (Nick collision from services.)
15:00:44rasherSomething good will come from cabbiev2 as default: theme authors will now be forced to think about setting background colors/backdrops
15:06:13gevaertsIt seems that my high speed reliability warning might be premature : I just checked the md5sums using the OF, and they fail as well, so the files on my sd card got corrupted (probably during some experiment).
15:07:01gevaertsConclusion: I don't know yet if high-speed corrupts data
15:07:35Leargevaerts: I read some 600 MB in high speed, and that seemed fine at least.
15:08:58gevaertsLear: that's good to know. I still get ehci errors in my dmesg output, so I won't enable it by default yet, but I still prefer instability to data corruption
15:09:40 Join qwedsa_ [0] (
15:10:09LearThat was on Windows XP, btw. And I only checked by playing some Oggs (and related album art files).
15:10:23DerPapstit is disabled currently in svn?
15:11:17gevaertsDerPapst: high speed is doubly disabled. You need USE_ROCKBOX_USB for full speed, and (additionally) USE_HIGH_SPEED for high speed
15:12:07rvvs89USB has strange naming conventions
15:12:18rvvs89If you can call them conventions
15:13:06gevaertsrvvs89: Doesn't the succession of low speed, full speed, high speed and super speed look well thought out and logical to you ?
15:13:47*gevaerts hopes not
15:13:47amiconnImho they are quite logical given how usb evolved
15:14:17gevaertsYes. You have to take historical context into account here
15:14:33BigBambiDear me, someone isn't happy:
15:14:46amiconnI'm not sure whether it was just a coincidence (because what stripwax reported earlier sounded like the same effect but with fullspeed), but windows doesn't seem to like rockbox highspeed mode
15:15:20amiconnIt slows down to a crawl (even lower than lowspeed), and sometimes accesses even seem to time out, causing windows to complain that xyz could not be written
15:15:28 Join TMM [0] (
15:15:45rvvs89gevaerts: Heh, I wasn't sure if you were being sarcastic or not, so I said nothing
15:16:06 Join jurrie [0] (
15:16:58gevaertsamiconn: linux isn't entirely happy about it either (I get ehci_hcd 0000:00:1d.7: dev 4 ep1in scatterlist error -104/-121 and lots of resets in the output), but it seems to recover faster (fast enough that the overall speed doesn't seem to suffer). I am now testing if the data survives
15:17:30rasheramiconn: Still, it's amazing that they never thought while developing usb 1.0 that they might come out with something faster at some point so using "full speed" was perhaps not the best idea
15:17:49 Part jurrie ("Leaving")
15:17:49amiconnWell, linux caches lots of data at the filesystem level. Try mounting with the sync option to make it act more like windows...
15:18:26gevaertsAnyway, I guess that 'FAT: Filesystem panic' answers my reliability question quite well
15:18:28dionoeaStill no rockbox people at FOSDEM?
15:18:40 Join midgey [0] (
15:18:43 Join webguest20 [0] (n=5b4ca4a4@gateway/web/cgi-irc/
15:18:47*amiconn doesn't like this "all or nothing" approach for caching filesystems in linux
15:18:55 Quit webguest20 (Client Quit)
15:20:22gevaertsrasher: I think they got some inspiration from radio, with VHF, UHF,...
15:21:22rvvs89However, radio never called anything "full frequency"
15:22:07*rvvs89 will stop being off topic
15:24:08*gevaerts has noticed that there's nothing like some slight off-topicness to see who's online and actually reading the conversation
15:25:01 Quit DerPapst ("plop")
15:25:38 Quit qwedsa (Read error: 110 (Connection timed out))
15:28:23rasheramiconn: I'm sure that there are kernel parameters to tune the caching strategy
15:30:32 Quit MajorC ()
15:32:14 Quit gevaerts (Remote closed the connection)
15:34:18 Join asdrubal [0] (
15:34:25asdrubalrbutil isn't detecting my C200
15:34:29asdrubalI got the 2.4 version
15:36:51 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
15:37:14stripwaxamiconn - right, I'm not using high speed
15:37:44*gevaerts just shut down his laptop by accident _yet again_
15:41:19 Quit hannesd (Read error: 113 (No route to host))
15:43:24 Quit midgey ()
15:43:35 Quit XavierGr (Nick collision from services.)
15:43:39gevaertshigh-speed reading seems to be fine. The 750 MB I read are correct
15:43:46 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
15:45:38stripwaxHm. Maybe I am using High Speed? Windows XP tells me "This device can perform faster if connected to a Hi-Speed USB 2.0 port"
15:46:21 Join carate [0] (n=terminal@
15:46:24gevaertsstripwax: That means you're using full speed. Windows checks the device descriptor USB version info, and if it sees 2.0 and a full speed connection it gives that warning
15:46:33stripwaxah, ok
15:46:41 Join Axio [0] (n=Axio``
15:46:57 Part carate ("Leaving")
15:47:16*gevaerts has to go for a few hours, but will be back later
15:47:19 Quit gevaerts ("bye")
15:47:42stripwaxSo I am using full speed, and Windows XP doesn't like it..
15:49:45 Join crzyboyster [0] (n=4b596e7a@gateway/web/cgi-irc/
15:50:02stripwaxRockbox is happily showing me the USB screen, but again I don't have a drive letter in Windows, and the new hardware detection is hanging again
15:50:28crzyboysterCan someone add the new cabbiev2 theme as default theme major change at >
15:50:49stripwax(which means I don't even have an option to 'safely remove' it)
15:50:54 Quit Xerion (Read error: 104 (Connection reset by peer))
15:51:10JamPSistripwax, it's buggy ok. you have to shutdown the player and then plug in the cable...
15:52:01 Quit crzyboyster (Client Quit)
15:52:48stripwaxJamPSi - not entirely sure what you are suggesting - that USB mode hangs if the cable is inserted while rockbox was already running? I merely booted with the cable plugged in
15:53:28stripwaxamiconn - is that consistent with what you were seeing? Do I really need to switch off the player and then insert the cable for it to work?
15:53:45JamPSiI dunno. I have to do it with my sansa now.
15:54:01JamPSiIt freezes if rockbox is running. Have to reset.
15:54:39stripwaxJamPSi - no, I am testing the rockbox USB stack with USE_ROCKBOX_USB enabled. So I need to have rockbox running :-)
15:56:59stripwaxok, windows is back, safely disconnect, remove cable -> data abort at 03f048f4 ..
15:58:39stripwaxhave to head off for today, happy to try suggestions & will check logs tomorrow when i'm back
15:58:50 Join midgey [0] (
16:00:07 Join Mathiasdm2 [0] (
16:00:39 Join Mathias [0] (
16:01:52 Quit qwedsa_ (Read error: 110 (Connection timed out))
16:02:00 Quit Mathiasdm2 (Read error: 104 (Connection reset by peer))
16:04:29 Quit asdrubal ("Leaving")
16:08:19 Quit Nevtus ("Gone")
16:08:28 Quit stripwax (Read error: 104 (Connection reset by peer))
16:10:58 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
16:12:33 Join Assdribbl [0] (
16:12:42 Nick Assdribbl is now known as Asdrubal (
16:13:21 Nick Asdrubal is now known as asdrubal (n=dribbles@unaffiliated/asdrubal)
16:14:53 Quit japc (Connection reset by peer)
16:16:54 Quit Mathiasdm (Read error: 110 (Connection timed out))
16:17:39*Hodapp examines increasingly shorter list of "Why default iPod firmware is good"
16:19:04Hodappwell, it appears in the absence of any other reason to want to use Rockbox, I'll use it because I'm too lazy to write a script to convert all my Vorbis files to some other format
16:19:56 Quit Nevtus ("Gone")
16:20:48 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
16:23:41 Quit Axio (Remote closed the connection)
16:36:42 Quit Nevtus (Remote closed the connection)
16:37:36 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
16:40:48 Join K4rP4D [0] (n=KrPD@unaffiliated/krpd)
16:43:21***Saving seen data "./dancer.seen"
16:52:49LearNico_P: Around?
17:04:18 Quit spiorf (Remote closed the connection)
17:06:14 Join hannesd [0] (
17:08:53Nico_PLear: yes
17:10:15LearTrying to debug some of the buffering issues, like showing playlist index 2 when playing track 1, and I've found one strange thing.
17:11:24Learaudio_fill_file_buffer buffers one track, then gets a next track request from the codec, as if it failed to decode the first one. This triggers a full rebuffer...
17:11:57Nico_Pthat's strange indeed
17:12:31LearCould be that album art delays the initial availability of file data, causing the codec to fail. I'm looking into what happens in the codec right now...
17:12:47Nico_Pthanks for looking
17:13:05Nico_PI've had little time on my hands lately, but I should have more soon
17:14:26LearA simple fix could be to set taginfo_ready to true once some file data is available (or whatever is needed to make the codec properly wait for data).
17:15:23 Quit desowin ()
17:15:35Nico_Psome time ago jhMikeS and I discussed the possibility of making the codecs use messages
17:17:43Nico_PLear: have you enabled logf output? and what are you working on (sim or target)?
17:18:54 Join desowin [0] (n=desowin@atheme/developer/desowin)
17:19:03LearWorking on target, as I only see it there. I have some logf yes, but with some extra output... (No queue messages though.)
17:19:37LearYep, codec fails to open the file.
17:23:32LearOr rather, initial read fails (returns zero), I think.
17:23:55Nico_PI think I saw a similar issue on the sim once. I was seeing Q_AUDIO_FLUSH messages appear twice
17:25:27LearBtw, possible bug: bufread can return < 0 on error, but codec_filebuf_callback doesn't check for that.
17:28:01 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
17:34:02HodappAnyone have a measure of normal battery life on a 3G iPod? I'm trying to figure out if this battery needs replacing or not... it was acting up a bit in the normal firmware, if I recall
17:34:23Hodappinfo says battery: 82%, 3h 30m
17:34:51LloreanIf you've never replaced it, it needs replacing.
17:34:59Hodappheh, that's what I figured
17:35:13LloreanLithium batteries don't age well, and 3Gs are quite old.
17:35:54Hodappalso, can I charge it while in Rockbox (with e.g. my AC adapter with a Firewire port) or do I need to go into Disk Mode or normal firmware for that?
17:37:00HodappI couldn't tell if it was charging or not when I tried
17:38:05 Quit corevette (Remote closed the connection)
17:39:36LloreanRight now it'll probably charge better in the Apple modes.
17:41:04gevaertsI don't know about the ipods, but on my sansa charging seems to stop too soon in rockbox
17:41:13HodappI suppose it will matter more when I have a battery that actually can charge properly.
17:41:46Hodappand even then it barely matters because it's just a matter of taking 10 seconds to reboot into a different mode
17:41:47 Part Llorean
17:43:03gevaertsdesowin: I am seeing the "shifted by two bytes" problem now on high speed, so I suspect it is real after all. Still no clue on where it comes from unfortunately
17:43:14 Quit barrywardell ()
17:44:14gevaertsdesowin: by real I mean a problem in the usb stack, and not some random bit-flip on our PC.
17:45:03 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
17:45:20desowinon bit-flip on your pc too ;) but yes, I understood what you meant
17:46:14gevaerts"Once is coincidence, twice is happenstance, three times is enemy action.", so let's wait until someone else sees it and then retaliate
17:46:37HodappStale quotes suck.
17:48:04 Join Stripwax [0] (n=Stripwax@
17:49:27gevaertsdesowin: by the way, what tools did you use to find the difference ?
17:50:15desowindownloaded shareware of hex workshop pro, I wished I found opensource tool for that...
17:54:32 Quit petur ("switching")
17:55:06*desowin is not keen on hacking khexedit
17:55:19 Join petur [50] (n=petur@rockbox/developer/petur)
17:55:35*gevaerts is now trying combinations of od and wdiff
17:56:27 Join webguest37 [0] (n=5b4ca4a4@gateway/web/cgi-irc/
17:58:54webguest37hello!i am porting rockbox to meizu m6/\.i need to to decrypt the original firmware (the code is in a rar archive which is in the firmware file after the first 8KB) here is how it should be done in M6.EBN chapter.but i dont understand how to do it.any ideas?
18:02:09 Quit Stripwax ("used jmIrc")
18:02:47LearNico_P: I'm not sure in detail what happens yet, but while bufopen is executing, the codec gets a chance to run, so the audio_hid is probably not ready by the time the codec tries to read data.
18:04:32 Quit midgey ()
18:05:13LearNico_P: Hm, should be safe to move the setting of taginfo_ready to after the audio_hid is set. Perhaps not the "right" fix, but should be good enough, I think.
18:05:31 Join BitTorment_ [0] (
18:06:53gevaertsDoes the SD card driver/spec work with 512-byte blocks internally, or can it do byte-addressing ?
18:07:29*gevaerts is trying to isolate this 2-byte data shifting bug somehow
18:08:45 Quit fyrestorm (Read error: 104 (Connection reset by peer))
18:10:52 Join webguest24 [0] (n=5b4ca4a4@gateway/web/cgi-irc/
18:10:52 Quit webguest37 ("CGI:IRC (EOF)")
18:11:34webguest24any ideas?
18:12:10 Quit Febs (Read error: 110 (Connection timed out))
18:13:15 Join fyrestorm [0] (
18:13:18gevaertswebguest24: from what I read there you need to remove the first 8 kB of the file, and the rest should then be a rar file. Any binary/hex editor should be able to do that
18:13:40 Join hannesd_ [0] (
18:13:55fyrestormsome of the newer builds of rockbox are causing my x64 vista to BSOD
18:15:02fyrestormusing rev 20080222 still works though
18:15:02fyrestormhappens when i plug it in
18:16:11fyrestormtried this on my other system, running vista 32
18:16:11fyrestormdidnt happen
18:16:11*gevaerts congratulates petur on his success, and now is curious what it was
18:16:11DBUGSent KICK fyrestorm to server
18:16:11fyrestormyes it did
18:16:11fyrestormjust tried it again
18:16:11***Alert Mode level 1
18:16:11DBUGsent MODE #rockbox +b *!*n=fyre@*
18:16:11fyrestormsystem is frozen
18:16:11*petur was just responding to yet another proof of 'vista sucks' - no local success here :/
18:16:13Kick(#rockbox fyrestorm :No flooding!) by logbot!n=bjst@gateway/web/cgi-irc/
18:16:13***Alert Mode level 2
18:16:13Mode"#rockbox +b *!*n=fyre@* " by logbot (n=bjst@gateway/web/cgi-irc/
18:17:31 Quit webguest24 ("CGI:IRC")
18:17:51 Join webguest13 [0] (n=5b4ca4a4@gateway/web/cgi-irc/
18:17:51webguest13thanks i will try it
18:18:04*petur has a chat with logbot
18:18:38 Quit BitTorment (Read error: 110 (Connection timed out))
18:19:34 Join barrywardell [0] (n=barrywar@
18:20:57 Quit EspeonEefi (Read error: 110 (Connection timed out))
18:21:11DBUGsent MODE #rockbox -b *!*n=fyre@*
18:21:16webguest13oh i quited irc.could you please repeat your message?
18:21:29Mode"#rockbox -b *!*n=fyre@* " by logbot (n=bjst@gateway/web/cgi-irc/
18:22:00 Join EspeonEefi [0] (i=espeonee@CERULEANCITY.MIT.EDU)
18:22:31gevaertswebguest13: the channel logs are available at
18:23:46 Quit EspeonEefi (Client Quit)
18:25:47webguest13<gevaerts>i think that couldnt help see i am from russia so i cant view logs for my time and date
18:26:14***Alert Mode OFF
18:26:41krazykitno, but surely you can figure out what time it is in UTC.
18:26:53gevaertswebguest13: why not ? just look at today's log and look near the bottom
18:29:02LearNico_P: I think that did it. I'll commit the small change soon.
18:29:02 Quit hannesd (Read error: 110 (Connection timed out))
18:29:02 Nick hannesd_ is now known as hannesd (
18:29:02webguest13and what date do you have there?we have 21 of february
18:29:02Nico_PLear: great :)
18:30:09gevaertswebguest13: There is a link marked "Today's log". That's the one you need
18:30:49Nico_Pgevaerts: have you tried cmp?
18:31:30 Join gregzx [0] (
18:31:41 Join fyrestorm [0] (
18:32:05gevaertsNico_P: cmp is not really useful to spot patterns. I'm now playing with od, wdiff, grep and others.
18:32:24*petur fails to find a logbot command to make it offer apologies to fyrestorm
18:32:58fyrestormcan anyone else confirm what i said though?
18:33:21peturjust don't write too many short lines in a short interval
18:33:21*alienbiker99 wishes he could get the gigabeat s bootloader installed
18:33:57fyrestormi replicate the problem on two different machines on vista. one 64bit, the other 32. has been happening with builds after 20080222
18:34:33Nico_Palienbiker99: what OS are you on?
18:34:33Nico_PjhMikeS: here?
18:34:33 Quit petur ("time to go, real life is calling...")
18:34:34alienbiker99and i have cygwin and vmware
18:34:59Nico_Palienbiker99: oh. sorry I'm afraid I can't help you then. I don't think sendfirm works in cygwin. it might be worth trying in vmware though
18:35:42 Join midgey [0] (
18:36:12 Quit jurrie_ ("Leaving")
18:37:07gevaertsfyrestorm: are you using official builds or your own builds ? Do you have the svn revision numbers ? I'm trying to correlate what you report with recent usb changes.
18:37:31fyrestormofficial builds.
18:37:56alienbiker99ah ok. thanks though Nico_P
18:38:29Nico_Palienbiker99: what have you been trying? the hacked V updater?
18:38:29webguest13<gevaerts>and then what shall i do with that .bin file from archive?
18:38:40alienbiker99no, ive been trying to get the sendfirm to work. i use my gigabeat s everyday so i dont want to flash it and have the hd wiped
18:38:48 Quit Zom (Remote closed the connection)
18:38:59 Join Zom [0] (
18:39:19jhMikeSNico_P: here here and there (checking in periodically)
18:39:21gevaertswebguest13: I can only read what that wiki page says. I'm not working on it myself.
18:40:04fyrestormgevaerts: rev 16365 is the last one that worked without a problem
18:40:04BigBambialienbiker99: masochist
18:40:09LearNico_P: As a nice side effect, effective buffering time is now noticeably shorter, as stuff isn't rebuffered. :)
18:40:10Nico_PjhMikeS: was wondering whether you had gotten sendfirm to work on windows (or with a VM). also thinking about your comment about the concurrency problems in the playback code
18:40:29alienbiker99i just cant get the libmtp to pick up the s in windows. i make the drivers and install it but it doesnt work
18:40:29jhMikeSyes, I have a sendfirm that works in XP
18:40:41Nico_PjhMikeS: good news? why no commit?
18:40:50BigBambialienbiker99: I mean using the gigabeast OF everyday :)
18:40:52Nico_Perr replace the first '?' by a '!'
18:41:03jhMikeSNico_P: It's messy to build and use
18:41:34alienbiker99BigBambi hahah yeah it sucks, but i have to. my h300 battery is dead =\
18:41:57BigBambibad luck
18:42:12jhMikeSIt needs alot of mods to libmtp though I could supply it compiled. It also needs driver installation to use it and then removal to access the MTP via explorer.
18:43:18alienbiker99i downloaded sendfirm from you via a link in the logs. after i install the drivers, how do i remove MTP access?
18:43:25***Saving seen data "./dancer.seen"
18:43:28 Quit webguest13 ("CGI:IRC (EOF)")
18:43:57Nico_PjhMikeS: a compiled version would be fine I guess. win users don't like to compile much ;)
18:44:13jhMikeSalienbiker99: just remove the libusb device in the device manager while the S is plugged and it will redetect as MTP
18:45:27 Quit nplus (Read error: 104 (Connection reset by peer))
18:45:32alienbiker99but thats after i install the bootloader correct?
18:45:55alienbiker99i cant get sendfirm to recognize the S
18:48:12jhMikeSyou did the device install with install-filter.exe?
18:48:31LearOK, committed. Time for dinner! :)
18:48:46alienbiker99hm no, i didnt
18:49:07jhMikeSit's in the .gz file in /bin
18:49:30 Quit midgey ()
18:51:06gevaertsCan anyone think of a mechanism that inserts (not replaces) two bytes on random addresses ending in 0x220, copying them from 32 bytes back ?
18:51:18 Join Buschel [0] (
18:51:54 Join midgey [0] (
18:52:46alienbiker99jhMikeS how do i use install-filter, through cmd?
18:53:04jhMikeSNico_P: btw, I've never had any extensive look at threading in playback since mob but I haven't notices is what prevents a handle from being moved or discarded while something still needs it
18:53:46*flyback is really getting fed up with firefox in general
18:56:02gevaertsfyrestorm: I can't really see any changes after r16365 that could cause that, but I can't test on vista here. That's ipod 5g, right ?
18:56:43jhMikeSalienbiker99: shi+...I'm trying to remember. there were some simple instructions in there iirc.
18:56:58jhMikeSoh's inf-wizard
18:57:01fyrestormgevaerts: correct
18:58:27jhMikeSalientbiker99: just run that with the S plugged and select it then it creates some driver .inf files. you use that to do a manual driver update on the MTP device from the device manager.
19:00:16alienbiker99ok ill try taht
19:00:55alienbiker99it still comes up as gigabeat S
19:01:24*flyback building a canucking jtag today )(*!*(!)(*#*(*)#@*(&*(&@##$(*)
19:01:36alienbiker99in device manager
19:02:11jhMikeSYou did Update Driver...?
19:03:06alienbiker99yes im trying it again
19:03:35alienbiker99samething. it still is being recognized as gigabeat s
19:03:40jhMikeSmake sure to install from a specific location and selecte the .inf inf-wizard created
19:04:31alienbiker99ok im gonna start over, unzip sendfirm and try again
19:05:19jhMikeSWhen I'm done I see LibUSB-Win32Devices/gigabeat S
19:05:35jhMikeSthen sendfirm will work
19:05:43alienbiker99and you see that in device manager?
19:06:26alienbiker99when im making the inf in inf-wizard. should it come up as MTP Device or gigabeat S?
19:06:43jhMikeSGigabeat S in the wizard
19:07:03jhMikeSmake sure the dll's are extracted too
19:07:25*flyback wonders if it's possible to port a mini os to the baseband processor on a server board
19:07:53alienbiker99the libusb0 dlls ?
19:08:06flybackyou know what might be a fun rockbox port
19:08:20flybackonly thing is you need a raw source of nand for firmware
19:08:21jhMikeSyeah, it needs those. I'd just extract the whole /bin directory and run inf-wizard from there
19:08:24flybackjuicebox port
19:08:41alienbiker99ok thats what im doing
19:08:45jhMikeSsave the inf the same one
19:09:07alienbiker99man its only coming up as MTP but with the same vendor id
19:09:41 Join EspeonEefi [0] (i=espeonee@STRATTON-TWO-EIGHTY-FOUR.MIT.EDU)
19:11:04alienbiker99jhMikeS do you have an updated sendfirm. maybe its my computer because it keeps recognizing it as a portable device
19:16:21 Join as [0] (
19:16:30jhMikeSsendfirm doesn't sound like the issue
19:16:59 Nick as is now known as as_ (
19:17:19 Part as_
19:17:38 Join ali_as [0] (
19:19:35jhMikeSI think I see what's going wrong though
19:19:59 Join dcbahr [0] (n=808a06bc@gateway/web/cgi-irc/
19:20:35jhMikeSmake sure to choose "Don't Search. I will choose the driver to install." and the click the "Have Disk..." button
19:21:13 Quit dcbahr (Client Quit)
19:22:15 Join BlakeJohnson87 [0] (
19:26:59 Quit XavierGr (Nick collision from services.)
19:27:09 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
19:30:37 Join barrywardell_ [0] (n=barrywar@
19:30:38 Join lee-qid [0] (
19:31:56 Join dcbahr [0] (
19:32:16 Join rocko [0] (n=rocko@
19:32:20dcbahrhi looking for tech help before purchasing an ipod
19:32:59rockoumm.. I got a question, my ipod (ipod color) and I were out all day yesterday (20 degrees)
19:33:10rockoand for whatever reason, all my fonts and whathaveyou disappeared
19:33:29krazykitrocko, that doesn't really sound related.
19:33:29rockoand it reverted back to the regular small text rockbox
19:33:41rockothat was the only thing I could think of
19:33:50krazykitdcbahr, the full list of supported devices is on the front page
19:34:09krazykitrocko, well, have you reloaded the WPS and font package, just to make sure everything's there?
19:34:24dcbahrI understand, but Ipurchased a 5th gen ipod vid last may and installed rockbox on it, but could barely get the thing to speak, I am tally blind so speech is essential,
19:34:39krazykitrocko, your "theme"
19:34:44dcbahrI am now looking at a 4th gen ipod photo but don't want to run in to the same issue
19:34:54rockoI don't know man
19:36:01krazykitdcbahr, every player has the same method for installing speech support.
19:36:03alienbiker99thats what i was doing wrong, thanks jhMikeS
19:37:16dcbahrok, perhaps I didn't install the speech support properly? this was several months back so I couldn't tell you exactly what I did, but I remember installing and uninstalling the firmware several times and just barely getting to to anounce a playlist, I never got any music out of the thing at all and returned the player
19:38:01 Quit BlakeJohnson86 (Read error: 110 (Connection timed out))
19:38:09 Quit barrywardell (Read error: 113 (No route to host))
19:38:10krazykitdcbahr, as i recall, it's as simple as downloading a file, renaming it, putting it into the proper directory and enabling an option
19:39:14dcbahrwhich iswhat I did and still got nothing, I really would like an ipod as mp3 cds take up space and this person has a good deal on one, could you possibly point me to the page for installing speech? so that when I get the player I can do it and try?
19:39:44jhMikeSa more minimal zip:">
19:40:17dcbahrthis was the latest and greatest ipod that I had purchased the last time, and I am thinking that the build had some bugs in it, so I'm wondering if an older model might solve the problem, not to mention it's cheaper
19:40:38 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
19:40:54alienbiker99jhMikeS i should be able to get into the default rockbox menu with the latest build right?
19:41:54jhMikeSIt should. Be sure to remove the libusb device from the device manager and do 'make tar' and copy that over using Explorer
19:42:14 Quit Mathias ("Yuuw!")
19:42:21alienbiker99so the .rockbox folder will be in a .tar?
19:42:49jhMikeSyes, and the bootloader will extract it.
19:43:00jhMikeSNico_P: iirc that is working now?
19:44:09krazykitdcbahr, , though i believe you rename the file to "english.voice" for an english voice.
19:44:22*jhMikeS hasn't tried the tar thing himself yet
19:44:37 Join criznach [0] (
19:45:08Nico_PjhMikeS: yes the tar thing works
19:45:17 Part criznach
19:45:50 Quit Lear ("ChatZilla 0.9.81 [Firefox 3.0b4pre/2008022204]")
19:46:08Nico_Palienbiker99: just send rockbox.tar over to the device (via MTP) and the rockbox BL will untar it
19:46:56dcbahrstrange, I rememberdoing this last year, the renaming and the langs directory, but still got no response, I did not have any sighted assistance though so it was all guessing
19:47:17alienbiker99ok thanks guys
19:47:23BigBambibluebrother: Sorry, I appear to have deleted something you werre replying to
19:47:28Nico_Palienbiker99: is it working?
19:47:48alienbiker99im building now, then making it a tar
19:47:51BigBambiBut I have tried to contain it all here:
19:48:23BigBambijhMikeS: The tar thing works excelently
19:48:29*BigBambi can't spell :(
19:51:46bluebrotherBigBambi: no biggie, I wasn't sure if I should submit that response anyway.
19:51:53alienbiker99i put the tar in root right?
19:52:18BigBambibluebrother: I still don't know what he is on about
19:52:19jhMikeSjust dump it in "Storage"
19:52:20Nico_Palienbiker99: yes
19:52:33BigBambiCan you put the tar anywhere else?
19:52:41*gevaerts is finding some more patterns
19:52:48bluebrotherI'd say he's trying to sell something he has absolutely no clue about. And wants us to fix that.
19:52:55BigBambigevaerts: You taken up knitting?
19:52:55jhMikeSIt only looks in the storage root
19:53:23BigBambibluebrother: Yes, it appears that way. I just want some details!
19:53:25gevaertsBigBambi: no, trying to find the cause of some data corruption
19:53:28Nico_PAFAIK the MTP DB puts all the files in the same dir anyway
19:53:48BigBambigevaerts: Good work :)
19:54:16gevaertsBigBambi: it's what makes high-speed unusable right now, but desowin has also seen it once with full speed
19:54:21*jhMikeS wonders what all the silly file munging accomplishes
19:54:50desowinso we have 3rd peron now!
19:55:26BigBambigevaerts: So I was lucky that high-speeed workjed for me when I tested it? :)
19:55:52jhMikeSgevaerts: is the data corrupted before or after sending to ATA?
19:56:12gevaertsBigBambi: did you read or write ? It goes wrong on write
19:56:32gevaertsjhMikeS: I don't know yet.
19:56:50BigBambigevaerts: Both
19:57:15 Nick fxb__ is now known as fxb (
19:57:20BigBambiBut I admit I didn't MD5, I just listened so there may be something wrong I didn't notice
19:57:54jhMikeSgevaerts: Sure there's no race condition with the interrupt? If anything must be atomic with it use set_irq_level. I'm guessing it may be more critical at high speed.
19:59:11 Quit ch4os_ (Read error: 110 (Connection timed out))
19:59:21gevaertsWhat happens is that "something" inserts two bytes at an offset of 0x220 into a page. The bytes are copied from 32 bytes earlier. Note that they are _inserted_, not overwritten. After this, the data is shifted by two bytes for a length of 16k-0x220 bytes.
19:59:24Nico_P"[18:53] <jhMikeS> Nico_P: btw, I've never had any extensive look at threading in playback since mob but I haven't notices is what prevents a handle from being moved or discarded while something still needs it" => handles are opened or closed only by the playback thread ATM, so there is no risk. There is nothing to prevent unwanted closing though. about moving, nothing prevents it either. the idea is that users should know the data can move and
19:59:24Nico_Ptherefor not rely on a pointer
19:59:34rasherWhich WPS-related changes (that would make a wps stop working) have there been since november? Trying to fix a wps (BigAqua for e200) for my father
20:00:11Nico_Prasher: maybe parameter syntax. run it through checkwps
20:00:31rasherNico_P: Dang - I don
20:00:38Nico_PjhMikeS: of course I welcome any improvement suggestions
20:01:02rasherNico_P: Don't have access to a linux machine here, and my own sims don't work nice with checkwps :-/
20:01:24jhMikeSbut if anything switches contexts in between for whatever reason the data consumer may have an invalid pointer
20:01:32gevaertsjhMikeS: that's a possibility. I haven't done much more than finding patterns yet. This data is written to RAM by the usb controller, which fires an interrupt on completion, after which the software writes it to flash.
20:03:48*gevaerts doesn't like this kind of bug
20:06:42 Quit flyback ("Ex-Chat")
20:07:22bluebrotherNico_P: btw, I had the impression that playback became buggier in the last month. Had a hang after skipping albums recently
20:07:36bluebrother(but I need to admit that I haven't investigated any further)
20:14:19rasherIf anyone's interested in my WPS woes, the offender was not ending %xl commands with a |
20:14:47 Join Febs [0] (
20:16:06Nico_Prasher: that's what I thought
20:16:35rasherLuckily the checkwps tool helped me
20:16:43Nico_Pbluebrother: there were lots of new bug reports lately. Lear committed a change that might improve things
20:16:59rasherI need to investigate what's up with my sims and debugwps sometime
20:17:11Nico_Pwhat's happenning?
20:17:19rasherNothing at all
20:17:36 Quit desowin ()
20:18:02rasherThe command just executes and returns to the prompt. I think it's something to do with being able to write to stdout - the regular sim build also forks and returns to the command-prompt immediately when run
20:18:09rasherThis is the crosscompiled windows-sim
20:18:24rasherDoesn't happen if you build in cygwin I think
20:19:18bluebrotherplayback stopped, but wps showed playing icon (i.e. no audio output, time display stalling at 0:00, statusbar icon "play")
20:19:43bluebrotherpausing and unpausing got me back into the game.
20:21:25 Quit dcbahr ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
20:25:06 Quit ol_schoola ()
20:27:19 Join CaptainSquid [0] (
20:29:41*Nico_P takes off
20:30:32BigBambiIl est un avion!
20:32:48bluebrotherBigBambi: you're really making it easy for that guy ;-)
20:34:02BigBambibluebrother: I don't know what else to say
20:34:11 Join kiddo [0] (n=terro@
20:34:40kiddowould you go somewhere where you fought with people would you talk to people who you fought with earlier? and you went out of site or made him/her out of sight?
20:36:24BigBambikiddo: If you would care to a) put that in English and b) make it relevant to Rockbox we would appreciate it. If not, go away
20:36:25krazykitkiddo, please read the channel guidelines. this is an on-topic channel, so please take non-rockbox chatter elsewhere
20:36:53kiddowtf ios rockbox
20:36:58kiddois it same as beat box?
20:37:16BigBambikiddo: Read
20:37:36kiddois it for linux or freebsd or mac or windows?
20:37:57BigBambiall of the above
20:38:03 Quit iamben (Read error: 104 (Connection reset by peer))
20:38:12BigBambiIt is for many portable audio players
20:38:20BigBambiand your OS is unimportant
20:38:49 Quit BlakeJohnson87 ("Leaving.")
20:39:53 Join BlakeJohnson86 [0] (
20:40:33 Quit tessarakt ("Client exiting")
20:40:36kiddofucking sucks
20:40:40kiddoprobably for women
20:41:09 Join donutman25 [0] (n=chatzill@
20:42:12*BigBambi wishes he had logbot poking powers
20:43:27***Saving seen data "./dancer.seen"
20:45:02kiddoi have huge nose or nose that sticks out really down so what? do i look like a rat or scary? like a jew? also i am Bengali origin and brown/dirty looking
20:45:15bluebrotherhmm, how can someone get those powers? Is there any documentation for it?
20:45:48krazykiti should think you just have to bother whoever runs it enough ;)
20:45:56Mode"#rockbox +o Febs " by ChanServ (ChanServ@services.)
20:46:08Kick(#rockbox kiddo :Febs) by Febs!n=chatzill@rockbox/administrator/Febs
20:46:30 Join kiddo [0] (n=terro@
20:46:31kiddoi have huge nose or nose that sticks out really down so what? do i look like a rat or scary? like a jew? also i am Bengali origin and brown/dirty looking
20:47:08Mode"#rockbox +b *!*@ " by Febs (n=chatzill@rockbox/administrator/Febs)
20:47:25*jhMikeS blesses kiddo and sells some rope too
20:47:37Kick(#rockbox kiddo :Febs) by Febs!n=chatzill@rockbox/administrator/Febs
20:47:50 Join dave_ [0] (
20:50:26dave_I don't know if anyone else has had trouble with this (couldn't find a bug report), but I'm having trouble with resuming playback after shutting down my Gigabeat F40
20:51:04Mode"#rockbox -o Febs " by ChanServ (ChanServ@services.)
20:51:57bluebrotherBigBambi: seems our guy is loosing his nerves.
20:52:01dave_when I resume playback, it will first play a snippet of a previous song in the playlist, then either play the song I stopped at a position further along than it was, or play another song further down in the playlist
20:53:58*bluebrother wonders why this guy is that upset
20:55:50bluebrotherdave_: does that happen on resume-on-startup or when resuming manually?
20:56:13dave_resuming manually
20:57:15dave_there's been all kinds of playback issues recently
20:57:17bluebrotherhmm, I haven't experienced this. But I also don't own a Gigabeat.
20:57:40bluebrotheryep, I notices some playback issues too. Just not this one.
20:57:41 Join bertrik [0] (
20:58:03bluebrotherI'm not sure if there already is a tracker entry for that. Quite possible.
20:58:28dave_I looked for about 5 minutes in the recent entries, but couldn't find one
20:59:42bluebrotherNico_P might be aware of it but I think he just left a while ago.
20:59:53dave_that's fine
21:00:08dave_I think I talked to him a little while ago about another playback issue I was having
21:00:23dave_maybe a week or so
21:02:02 Join Sivart0 [0] (
21:03:26 Join Arathis [0] (
21:05:19dave_a ha!
21:05:19 Quit BlakeJohnson86 (Read error: 104 (Connection reset by peer))
21:05:26dave_I think I found what I was looking for
21:05:39 Quit donutman25 ("ChatZilla 0.9.81 [Firefox]")
21:05:47dave_well, for one of the issues, at least
21:05:55bluebrothernice :)
21:07:47jhMikeShmmm...codec_track_changed called from FIQ/DMA interrupt context is unsafe
21:08:54 Quit Sivart0 ("ChatZilla 0.9.81 [Firefox]")
21:10:22 Join ch4os_ [0] (n=ch4os@unaffiliated/ch4os/x-059673)
21:15:10 Join Domonoky_ [0] (n=Domonoky@
21:15:10 Quit Domonoky (Read error: 104 (Connection reset by peer))
21:15:32 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:17:24 Quit Nevtus (Remote closed the connection)
21:17:49gevaertsIs memcmp supposed to work on rockbox ?
21:19:04gevaertsThen I'm doing something stupid somewhere.
21:19:13 Join Guest38679 [0] (n=galgorha@
21:19:14Bagderlots of code uses it already at least
21:19:22 Join caysee [0] (
21:20:05*gevaerts found the stupidity, and tries again
21:20:50 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
21:22:50*gevaerts feels slightly relieved. The bug is not in the usb driver, but seems to be in the sd card driver
21:23:02gevaertsSo it's not entirely my fault at least.
21:23:58gevaertsIs ata_write_sectors() supposed to be interrupt safe ?
21:25:16gevaertsHow high do I need to set_irq_level() ? Disable ?
21:25:47jhMikeSyou can't call that function from an interrupt handler at all
21:26:04jhMikeSset_irq_level(HIGHEST_IRQ_LEVEL) disables everything but audio interrupts
21:26:24jhMikeSit returns the old level which can be restored by calling the same
21:27:44gevaertsI'm not in an interrupt handler. I get usb data, tell the controller to expect more, and call ata_write_sectors(). (probably) during the ata_write_sectors() call an USB interrupt comes in which seems to disturb the write to flash. reading back immediately afterwards shows the two-byte shift
21:28:30Febsdave_: I just came across this thread in the forum that seems similar to what you described earlier:
21:28:43bluebrotherDomonoky_: is complete install considered to be fixed?
21:33:38 Join jurrie [0] (
21:33:41dave_thanks Febs
21:34:25dave_looks like Marcus committed something today as an attempt to fix some of the playback problems
21:34:26jhMikeSgevaerts: the interrupt during a disk write shouldn't disturb anything. you're setting it up to receive another block into a different buffer while writing the last one received?
21:35:12gevaertsjhMikeS: yes.
21:35:17dave_I'm in the process of updating my build to check it out
21:36:31 Part Guest38679
21:36:50gevaertsjhMikeS: for testing purposes I'm reading back immediatelly what was just written, and it (sometimes) doesn't match, in exactly the right way (i.e. a two-byte shift)
21:36:58Domonoky_bluebrother: i consider it fixed, until i hear otherwise.. :-)
21:37:01jhMikeSgevaerts: I can't see a problem unless the interrupt could occur during setting up the next buffer by being setup to fire too early or some such.
21:38:22bluebrotherseems like some guy has problems using it on h10:
21:41:09 Quit bertrik ("bye")
21:42:39jhMikeSwhere is usb_drv_recv implemented? (can't seem to find it)
21:42:42Domonoky_bluebrother: this guy seams to have problems doing anything..
21:43:01gevaertsjhMikeS: in firmware/target/arm/usb-drv-pp502x.c
21:43:40*gevaerts still has his money on the sd card driver doing the wrong thing for some reason
21:46:37 Join BlakeJohnson86 [0] (
21:46:47jhMikeSso no interrupt can fire until REG_ENDPTPRIME |= mask ?
21:47:40gevaertsjhMikeS: I put a patch for my extra checking code at, so you can see how I check for the bug
21:48:33linuxstbali_as: I've just added your Samsung YP-K3 line to tcctool in SVN.,,
21:49:14gevaertsjhMikeS: not for this transfer at least. There could be other transfers pending at the same time if e.g. serial is active.
21:49:31 Join corevette [0] (
21:49:34 Quit dave_ ("Ex-Chat")
21:51:17linuxstbBagder: Are you around for a few minutes? I'm creating a new win32 binary for tcctool for the download server.
21:51:25BagderI am
21:53:12*jhMikeS still is unfamiliar with how everything is getting signalled here
21:54:04bluebrotherBagder: btw, have you gotten around moving the updated linux binary of rbutil to the download server? My first binary had libusb dynamically linked.
21:54:29Bagderlinuxstb: should they perhaps get a version number?
21:54:45Bagderbluebrother: no, I haven't seen any updated ones. URL?
21:54:52Domonoky_bluebrother: could you also update the windows binary again ? (there was a bugfix for the sapi_voice.vbs script.. )
21:55:29linuxstbBagder: Hmm. I'm not sure it's worth it - the only thing that changes is that more supported devices are added, and the user can see that list by running tcctool without any options.
21:55:53bluebrotherDomonoky_: the too-small-voicefiles issue?
21:56:06Bagderlinuxstb: ok, it's been updated now
21:56:19linuxstbBagder: OK, thanks.
21:56:34Domonoky_bluebrother: yes, it was non-working voice selection..
21:56:51 Join Rondom [0] (
21:57:14Bagderbluebrother: done!
21:57:20jhMikeSgevaerts: what do you mean a two-byte shift?
21:57:20bluebrotherBagder: thanks :)
21:58:45 Quit linuxstb ("Leaving")
21:59:16bluebrotherDomonoky_: ah, you're referring to r16359?
22:00:16Domonoky_bluebrother: i think you can just rebuild the m1.0.4 branch, because i only branched the rbutil dir, and the script is in the tools dir :-)
22:00:54gevaertsjhMikeS: at a certain point two bytes are written that shouldn't be there (they seem to be copied from 32 bytes earlier), and then the needed bytes are written, without any missing, so they are all in the wrong position.
22:01:03bluebrotherwell, we haven't had any changes to rbutil after the last release so it won't make any difference ...
22:01:07*gevaerts doesn't really know how to say this clearly
22:01:11barrywardell_bluebrother, Domonoky_: speaking of rbutil. I tried to build it on my Mac today. I got it to build, but it segfaults when going into the config screen. Any ideas why?
22:01:22bluebrotherexcept for the net traffic (which currently still _is_ an issue for me)
22:02:06barrywardell_here are the details of the crash in case it helps:
22:02:10Domonoky_and yes its rev16259..
22:02:22Domonoky_s/2/3 :-)
22:03:45 Join webguest12 [0] (n=5b4ca4a4@gateway/web/cgi-irc/
22:03:50bluebrotherbarrywardell_: do you have an old configuration file around?
22:04:14barrywardell_I did, but I deleted that and still get the same problem.
22:04:48bluebrotherhmm. Seems it has problems with setting the proxy values.
22:05:05 Join desowin [0] (n=desowin@atheme/developer/desowin)
22:05:18barrywardell_I deleted the config file in ~/.config/
22:05:23*Domonoky_ sees the problem..
22:05:38 Join robin0800 [0] (
22:05:38Domonoky_function systemProxy in utils.cpp
22:06:05Domonoky_it only has an implementation for linux and windows, no return value for mac .. :-/
22:06:30 Quit K4rP4D ("Leaving")
22:06:32barrywardell_ah, makes sense
22:06:43webguest12hello!i am porting rockbox to meizu m6.i need to decrypt the original frimware.i have a code in a .rar archive which contains .bin should i do it?
22:06:48*bluebrother just found that too
22:06:57*bluebrother blames Domonoky_ :P
22:07:11*Domonoky_ hides..
22:07:30jhMikeSgevearts: so it's particular bytes from the last 32-byte block or just random leaking of two bytes into the start of the current block?
22:08:05Domonoky_bluebrother: any idea how to get the systemproxy on mac ?
22:08:27 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
22:08:41*jhMikeS thinks he saw this stuff when attempting to speed up the writing in general
22:09:17gevaertsjhMikeS: in the cases I checked, the two "extra" bytes are always inserted just before a new 32-byte block, and are the same as the first two bytes in the preceding 32-byte block
22:10:01bluebrothersystem proxy selection is disabled on mac anyway (until now)
22:10:26gevaertsjhMikeS: is there any point in trying without boosting ?
22:10:27Domonoky_so a quickfix would be to return an empty url ?
22:10:54jhMikeSgevaerts: I think that should be tested to see if it goes away
22:11:33bluebrotheryep, just return QUrl("");
22:11:48bluebrotherbut I'm wondering why this function is called at all
22:11:48*barrywardell_ is testing that now
22:12:27barrywardell_it fixes the crash anyway
22:12:52*bluebrother found another bug
22:13:00barrywardell_in the config dialog, "Use system values" is greyed out, but selected
22:13:19 Quit robin0800 (" HydraIRC -> <- Nine out of ten l33t h4x0rz prefer it")
22:13:36 Join robin0800 [0] (
22:13:54webguest12or when i deleted first 8 kb from the frimware using hex editor and had a .bin file, i did decryption?
22:14:27bluebrothersystem proxy is broken in rbutilqt.cpp
22:14:48Bagderwebguest12: I think that's what is meant yes, as I believe there is no encryption involved
22:16:48Domonoky_also its the wrong default, if it isnt aviable on mac... should set the default to "none"
22:17:40webguest12and what is encryption and how will i know if it is involved or if it isnt?
22:18:55gevaertsjhMikeS: if I turn off boosting, it seems high speed doesn't work any more
22:18:56Bagderwebguest12: if you can extract the file(s) in the rar file and disassemble the arm code, there is no encryption
22:19:05bluebrotherI'm quite sure the default was none on mac.
22:19:26bluebrotherSeems we overlooked some small regressions during code refactoring :/
22:19:28*Domonoky_ is sorry for breaking bluebrothers proxy code... but i had to rework the settings... :-)
22:20:39barrywardell_I just deleted my config file and checked again. "Use system values" is selected and greyed out by default on mac
22:21:03 Join Xerion [0] (
22:21:37webguest12so if i can extract this file and i can open it with hex editor than there is no encryption?
22:22:14Domonoky_yes, again a little mistake of me, while i reworked the settings code (defaults are now in rbsettings) .. * me goes hiding* ...
22:22:16 Quit mf0102 ("Verlassend")
22:22:53Bagderwebguest12: if you can extract it, and read sensible data from the files, then yes
22:22:57 Join replix [0] (
22:23:15Domonoky_bluebrother: should i start fixing, or are you already orking on fixing my mess ? :-)
22:23:15replixwhere to get the firmware-roms for the sansa e200 models?
22:23:36webguest12so then what arw the second steps in porting?
22:24:13*barrywardell_ can build a rbutil binary once the fixes are committed to svn
22:25:24bluebrotherDomonoky_: well, I already committed a fix. But not for barrywardell_'s issue
22:27:11bluebrotheroh, something completely different: can someone explain me the syntax in apps/player/icons.c:42 (the ... part)? Is that an extension?
22:27:25*Domonoky_ commited the fix for barrys issue now..
22:28:08 Quit CaptainSquid ("Miranda IM!")
22:28:43 Quit replix (Read error: 104 (Connection reset by peer))
22:30:00webguest12so now i need just to try if the player works with that .bin file.if it is so then how to upgrade my player with this binary?(usually i just push to files m6.ebn and resource.bin in root
22:30:48 Quit jhulst (Remote closed the connection)
22:31:03 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:31:37Bagderwebguest12: you need to tell us, it's you who owns this player not us
22:31:52*amiconn is now really pissed by rockbox' current resume behaviour on swcodec
22:32:39amiconnResuming into the last track of an album practically never works, even if track would still have minutes to go
22:33:34webguest12but the second step is to check if the player works with that .bin file, right?
22:33:50jhMikeSgevaerts: not sure if it's if any use but a queue can be read by an interrupt handler and replies received to waiting threads as long as queue_wait_w_tmo is used in the handler and it's dedicated to thead to interrupt messaging
22:34:13jhMikeSof course the timeout should be zero to avoid blocking
22:34:54*bluebrother committed a fix for Domonoky_'s fix :P
22:35:18amiconnDomonoky_: Did you check whether my sapi5 list fix helps?
22:36:53gevaertsjhMikeS: I think I understand what you're saying, but I don't really see a need for it within the usb code.
22:37:58 Quit webguest12 ("CGI:IRC")
22:42:20 Join replix [0] (
22:42:28bluebrotherhmm, does w2k don't ship with sapi tts?
22:42:54amiconnIirc it ships with a *broken* sapi4 installation
22:43:31***Saving seen data "./dancer.seen"
22:43:39bluebrotherurgh. Ok, then I won't try building voices on that box.
22:43:46amiconn(I've read that somewhere - I don't have much experience with w2k myself)
22:44:03amiconnJust install sapi4 and sapi5. Then it should work
22:44:30bluebrotherwhere can I grab those? /me has no idea about this stuff
22:44:35amiconnRemember to register the sapi4 DLLs manually - the standalone sapi4 installer forgets that too
22:44:40jhMikeSgevaerts: Just a thought I had. Anyway, I did get that stuff before from the ATA and never found a logical reason for the stale halfword and nothing seemed to fix it but keeping writing slow.
22:45:06gevaertsjhMikeS: any way of slowing it down more ?
22:45:30jhMikeSlonger udelay in ata_write_sectors perhaps
22:46:15*barrywardell_ has a the osx version of rbutil m1.0.4 ready for the download server
22:46:24amiconnjhMikeS: Do I remember correctly that the sansa OF writes the last 16 bytes (or so) of each sector without using dma?=
22:46:31jhMikeSit's almost like FIFO_EMPTY isn't telling the truth
22:46:52amiconnMaybe that's some kind of workaround for (yet another) PP hardware bug....
22:46:56jhMikeSamiconn: there is that but I don't recall how much it writes
22:47:14jhMikeSwe aren't using any DMA at all anyway
22:47:46amiconnMaybe that's the problem...
22:48:27amiconnI remember a problem on Ondio, which had a different effect but also had to do with write latency
22:48:41jhMikeSI think that may be related to cache lines
22:49:03amiconnThe Ondio MMC driver *does* use DMA, transferring to/from SPI
22:49:53amiconnThere are 4 DMA channels in the SH1, with different priority. We used DMA1 for MAS audio DMA, and DMA2 for MMC DMA (lower priority than DMA1)
22:50:19amiconnSometimes, the MMC driver just locked up. Happened only on write, never on read
22:50:23jhMikeSthe doc I have just says "a small part of transfers using PIO"
22:50:25gevaertsjhMikeS: I now have udelay(5) instead of udelay(2), and it still goes wrong
22:50:48 Join Zagor [242] (
22:51:12amiconnIt also happened more often when playing music, and became very apparent with the WAV plugin (with of course puts more strain on both DMAs)
22:51:29jhMikeSmore accurately: "...a small part of every write using PIO"
22:52:13 Part Buschel
22:52:24barrywardell_Bagder: if you're around, can you please put this in the rbutil/macosx/ dir on the download server:">
22:52:45amiconnSo I guessed that the MMC DMA might get starved sometimes. I changed it to using DMA0 (higher priority), and the problem went away...
22:53:48bluebrothergtg now. bye.
22:53:52 Quit bluebrother ("leaving")
22:54:34amiconnjhMikeS: I guess it's starvation and not cache related because gevaerts reported that the following data is shifted
22:54:47jhMikeSamiconn: but what we have here is 2 bytes duplicated from the start of a previous FIFO fill as the start of the next. iow a slot never seem to get updated.
22:55:42 Quit robin0800 (Read error: 104 (Connection reset by peer))
22:56:05amiconnI guess this is because the controller continued writing although the fifo was already empty. The 32 bytes would then be the fifo length
22:56:15jhMikeSgevaerts: the IMX.31 SDHC controller is similar (though not identical) to the PP one so perhaps check that out
22:57:03jhMikeSthe card could be in the wrong state
22:58:11 Join Axio [0] (n=Axio``
22:59:07gevaertsjhMikeS: now that you mention it, the svn usb code still has ata_enable(false) somewhere in its path. I'll see removing that changes anything
23:00:53gevaertsjhMikeS: no change
23:01:08jhMikeSor just disable the function body in ata-c200_e200.c :)
23:02:03gevaertsI don't think I have mentionned this before, but it seems to happen about once every 1.5 MB
23:02:59jhMikeSI don't think you did. When I tried optimizing the write code it happened rather immediately. I wonder if recording would also show this problem.
23:04:48 Join robin0800 [0] (
23:05:36barrywardell_wouldn't test_disk reveal if there was a problem like that?
23:07:19gevaertsI'll try
23:07:53jhMikeScopy a huge file on target and compare it?
23:09:06gevaertsI'll start with the test_disk write and verify test, boosted and unboosted
23:10:30*gevaerts thinks people left some bugs around to trick unsuspecting contributors into improving parts of the code they weren't originally interested in
23:10:52 Quit jhulst ("Konversation terminated!")
23:11:09 Join Rincewind [0] (
23:12:42*jhMikeS has a similar sense at cynicism at times :p
23:12:52 Quit csc` (Connection timed out)
23:14:28*amiconn joins in
23:14:40 Quit Arathis ("Bye, bye")
23:15:54 Join csc` [0] (n=csc@archlinux/user/csc)
23:16:46krazykitgevaerts, you've uncovered the conspiracy plot :o
23:17:15*gevaerts hides before the Swedes discover this
23:19:18 Quit jurrie ("Leaving")
23:20:03gevaertsIt's lucky I just read a few days ago
23:20:58gevaertsWrite&verify at 30 MHz passed, now running at 80 MHz
23:21:37 Quit Rondom (Connection timed out)
23:24:21 Join jurrie [0] (
23:24:29gevaerts80 MHz passed as well.
23:24:38 Join jurrie_ [0] (
23:31:16 Join sdfsf [0] (
23:31:54 Quit replix (Read error: 104 (Connection reset by peer))
23:33:04 Quit Rob2222 ()
23:33:27 Join Rob2222 [0] (
23:34:11 Quit ch4os_ ("Lost terminal")
23:34:38gevaertsI copied a 25MB file from internal flash to SD card, once at 30 MHz and once at 80 MHz. The md5sums match.
23:36:09 Quit fyrestorm (Read error: 113 (No route to host))
23:46:14 Join herrwaldo [0] (
23:46:37 Quit waldo (Read error: 113 (No route to host))
23:47:37barrywardell_gevaerts: does the problem only happen with sd based devices?
23:48:20*amiconn now found the real cause of the failing default viewer iconset on greyscale targets
23:49:07gevaertsbarrywardell_: no idea. I only have a c250, and although desowin saw it once on full speed, it seems to happen nearly only on high speed connections, and that's not very widely tested yet
23:50:17gevaertsbarrywardell_: but I strongly suspect it happens if the sd driver is interrupted at a bad moment during writing.
23:51:26 Quit TMM ("Ex-Chat")
23:52:18barrywardell_I can test on my H10
23:52:47*barrywardell_ has found that he can use rockbox usb mode on his mac by running windows through parallels desktop :)
23:53:59gevaertsI just changed the udelay(2) in ata-c200_e200.c to udelay(1), and that seems to solve the issue (or at least make it rare enough that I didn't see it in my 100MB test-copy)
23:54:13jhMikeSgevaerts: what would interrupt it?
23:54:26midkayBagder: still around?
23:55:03gevaertsjhMikeS: the usb transfer completion interrupt for the next chunk of incoming data
23:55:28 Join aliask [0] (n=chatzill@rockbox/developer/aliask)
23:55:50*gevaerts goes to do some maths to verify his claim
23:56:05jhMikeSgevaerts: it gets interrupted by the tick too 100 times/second
23:56:34gevaertsjhMikeS: you have a point there
23:58:04jhMikeSit gets interrupted during playback 3675 times/second to fill audio fifos too :)
23:58:04barrywardell_what's the best way to test it on my H10?

Previous day | Next day