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

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

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

#rockbox log for 2008-12-03

00:00:20jeffroniusnice bob dobbs photo!
00:01:00 Quit toffe82 (Remote closed the connection)
00:02:04 Quit jerryluc ("bla bla woff woff")
00:08:07 Quit bertrik ("Leaving")
00:10:28 Join casainho [0] (n=chatzill@
00:12:22 Quit ender` (" A computer program will always do what you tell it to, and seldom what you want it to.")
00:13:25 Quit jeffronius ("KVIrc 3.2.6 Anomalies")
00:14:27 Join massiveH [0] (
00:15:29casainhohello :-) −− I am trying putting kernel working, but my system is hanging :-(
00:15:37casainhoon this code:
00:16:35 Quit casainho (Read error: 104 (Connection reset by peer))
00:17:03 Join casainho [0] (
00:17:08casainhoif I uncomment that line, the init_kernel()
00:18:20Bagderthere's no such in that file
00:18:26casainhocan someone please give me advice of why my code hangs?
00:19:35casainhoBagder: can you hep me?
00:19:41casainhohelp me?
00:20:43Bagdersure, show us a diff against rockbox
00:21:03casainhoI can't, I don't have it
00:21:25Bagderthen you're not really trying to get our help
00:22:09Bagder*everyone* who works on rockbox, new ports, new plugins and everything, works with patches against rockbox svn
00:22:09casainhoBagder: my question is that I have 2 SVN
00:22:11 Quit domonoky (Read error: 104 (Connection reset by peer))
00:22:35 Quit bmbl ("Woah!")
00:22:43casainhohow can I have the same code on two different SVN servers?
00:23:20Bagderthat's a totally different issue, and you won't have your code in the official svn repo until a committer would commit it there
00:23:30Bagderand that will happen once you post fine patches
00:24:16Bagderso you can have your code wherever you want, but when you want the code to get looked at and tested by rockbox hackers, you need to realize that patches against rockbox svn is the way forward
00:24:19casainhountil then, we are using another SVN server, so we can work, we are 3 guys
00:24:31BagderI'm not arguing against that
00:25:20casainhodo you know a way I can have the same code on SVN on and at local rockbox?
00:25:55Bagdersure, checkout from rockbox and commit to the google svn
00:26:12Bagderbut I thought the point is that you don't have *the same* code
00:26:52 Join toffe82 [0] (
00:27:01casainhoI can't checkout the same code to google SVN because there are .svn dirs from SVN of RB, I do not work that way...
00:27:16Bagderyou can use cp
00:27:25bluebrotherso what? co Rockbox svn, export the co, then cp to the other folder
00:27:48Bagdercode and diffs are just text files
00:29:21casainhoI will ask to other developer to try to make a script for do that, I am not able
00:29:24 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
00:29:31casainhothanks :-)
00:30:07Bagderbut what rockbox version are you using then?
00:30:31BagderI mean, what are you basing your port on
00:31:15casainhoon an older version... I just chekedout one time and then I started using that code...
00:31:34Bagderrockbox moves you know
00:31:58casainhomy question is that my code is hanging when I call a function...
00:32:34Bagderyes, and then I want to know how some of your files differ from rockbox's original ones
00:32:34 Quit casainho (Read error: 104 (Connection reset by peer))
00:32:57 Join casainho [0] (
00:33:05Bagderyes, and then I want to know how some of your files differ from rockbox's original ones
00:33:11Bagderhence my questions about diffs
00:33:19casainhoI don't really understand why, I already try to call another to test and it don't hang... also I had increaded the stack to 2Mbytes
00:33:27Bagderand now you say you use an old version, so you could thus suffer from old bugs
00:34:18casainhoI have memory for that.. also all looks perfect on memory map file...
00:34:33Bagderso does 2mb buffer solve anything?
00:34:41Bagder2mb stack I mean
00:34:47casainho2mb of buffer solves nothing :-(
00:34:50bluebrotherif you have a much bigger stack consumption than stock Rockbox something is definitely wrong
00:34:52Bagderhm MB it should be
00:35:16casainhono, I just defined thet stack size on linker sript to be 2MB
00:35:18Bagdercasainho: then I would say the stack is not the problem
00:35:49Bagderbad sdram config?
00:35:52casainhoyes, I am think stack is no problem... but is strange that calling that function it hangs...
00:36:05bluebrotherwhat does "this" function do?
00:36:10casainholooking at memory map, no..
00:36:28Bagdersdram config isn't seen in the memory map...
00:36:39BagderI mean refresh rates etc
00:36:50casainhothe function is this one: add_to_list_l(&cores[core].running, thread);
00:37:45casainhohmmm... all the code is working if I do not call that function, and SDRAM is configured from original code from Atmel...
00:38:14Bagderdoes it freeze on the actual branch instruction?
00:38:38Bagderit was somewhere within that function, wasn't it?
00:38:54casainhoyes, somewhere inside...
00:39:08jhMikeScasainho: is your BSS properly zeroed?
00:39:15casainhothe funciton is this one: static void add_to_list_l(struct thread_entry **list,
00:39:17casainho struct thread_entry *thread)
00:39:17Bagderand all the pointers and memory look fine ?
00:39:30casainholine 1251:
00:39:32Bagder(assuming you can debug over the jtag)
00:40:03casainhoI am not able to use JTAG... I coudn't make it working it, I am using a serial debug port
00:40:06jhMikeSthat won't fail unless the pointers are bad (it's just adding to a linked list)
00:40:54casainhojhMikeS: I don't know the answer...
00:40:59 Quit __lifeless (Remote closed the connection)
00:41:22casainhojhMikeS: so, why could make the pointers bad?
00:41:38linuxstbDoes your startup code initialise bss to zero?
00:41:56linuxstbAnd the other inits the Rockbox crt0.S files normally do?
00:43:43casainhothe linker script is this one, I am not sure it initialise the bss to zero:
00:43:58 Join _lifeless [0] (n=lifeless@
00:44:09bluebrotherthe linker script never initializes any sections. Your startup code has to do this
00:44:19bluebrotherusually this file is called crt0.S
00:46:54***Saving seen data "./dancer.seen"
00:48:16casainhohmmm, don't know :-)
00:48:23jhMikeScasainho: do enable the panics in kernel.c and thread.c and it will catch things like that
00:48:26casainhoif you can give a look:
00:49:56casainhopanics? it write messages to LCD, no?
00:50:50casainhoI don't have LCD :-)
00:51:03casainhohmm... but I have serial debug port...
00:51:04jhMikeSI see (or don't)
00:52:03 Quit jfc (Read error: 131 (Connection reset by peer))
00:52:18 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111318]")
00:52:27 Join jfc [0] (
00:52:38 Join casainho [0] (n=chatzill@
00:53:06casainhobluebrother: yes, the crt0.S file clears bss
00:53:18 Quit massiveH ("Leaving")
00:55:34 Quit JdGordon (Remote closed the connection)
00:55:55 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
00:58:54 Quit culture (Read error: 110 (Connection timed out))
00:59:47 Quit Nico_P (Remote closed the connection)
01:00:18 Quit herrwaldo ("Konversation terminated!")
01:01:45 Quit bluebrother ("sleep")
01:02:17 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111318]")
01:03:59 Quit _lifeless (Remote closed the connection)
01:04:03 Join __lifeless [0] (n=lifeless@
01:06:47 Quit Rob2222 (Read error: 104 (Connection reset by peer))
01:08:35 Join Rob2222 [0] (
01:10:12 Quit MethoS- (Read error: 104 (Connection reset by peer))
01:13:55 Quit J-23 (Read error: 60 (Operation timed out))
01:16:56 Quit Rob2222 (Read error: 104 (Connection reset by peer))
01:17:11 Quit tyfoo (No route to host)
01:17:57 Join Rob2222 [0] (
01:22:22 Join mikem [0] (n=mikem@unaffiliated/mikem)
01:23:41mikemhi, currently when i plug in my iPod's USB cable, I have to hold down Menu in order for the computer not to mount it. is it possible to swap this functionality, so that holding down menu is required for the computer to mount the iPod?
01:26:07linuxstbNo, there's no setting for that.
01:29:22 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
01:31:48 Join Rob2223 [0] (
01:32:25 Part toffe82
01:34:03 Quit moos ("Rockbox rules the DAP world")
01:34:36 Quit Rob2223 (Read error: 104 (Connection reset by peer))
01:39:25 Join Rob2223 [0] (
01:43:32 Join jgarvey [0] (
01:45:25 Quit faemir (Remote closed the connection)
01:47:40 Quit jhulst (Remote closed the connection)
01:51:30 Quit Rob2222 (Read error: 110 (Connection timed out))
01:51:50 Quit Rob2223 (Read error: 104 (Connection reset by peer))
01:54:34 Join Rob2222 [0] (
01:57:18 Quit lasser (Read error: 110 (Connection timed out))
02:00:18 Quit Rob2222 (Read error: 104 (Connection reset by peer))
02:00:33 Join Thundercloud [0] (
02:06:01 Join Darksair [0] (n=user@
02:09:06 Join mc2739 [0] (
02:10:44 Nick fxb is now known as fxb__ (
02:11:01mc2739is there a dev available to change
02:11:25mc2739it should be task type patches instead of bugs
02:11:54LambdaCalculus37mc2739: I'll fix that.
02:12:06 Quit mcuelenaere ()
02:12:17mc2739LambdaCalculus37: thanks
02:13:56 Quit Thundercloud (Remote closed the connection)
02:16:02 Part pixelma
02:16:15 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
02:24:54 Join Rob2222 [0] (
02:28:49 Join Gareth [0] (
02:33:24Unhelpfullatest scalers at FS #9458. there's a lot working, but a few odd issues that might need more eyes.
02:33:51 Join EspeonEefi [0] (i=eefi@STRATTON-SIX-FOURTY-SIX.MIT.EDU)
02:37:21 Part mikem
02:41:21 Quit Rob2222 (Read error: 110 (Connection timed out))
02:46:55***Saving seen data "./dancer.seen"
02:51:18 Quit obo ("bye")
02:51:28LambdaCalculus37Ladies and gentlemen, I have here an iPod video successfully modded with a CF card and an iFlash adapter. :)
02:52:19 Quit jgarvey ("Leaving")
02:53:43 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
02:54:52 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
02:55:11 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
03:10:22 Join nuonguy [0] (
03:11:56LambdaCalculus37Bah... spoke too soon... ATA Error -1. :/
03:18:19 Join tsfriendship [0] (n=tsfriend@
03:20:43Dhraakellianquestion: how much effect would the presence of a microSD card have on an e200's battery life?
03:21:58 Part tsfriendship
03:23:21 Quit alexbobp (Read error: 54 (Connection reset by peer))
03:25:12 Join krazykit` [0] (
03:26:05 Quit Darksair ("Do you hear that? This is the sound of inevitability. This is the sound of your death, Mr. Anderson.")
03:26:06 Quit krazykit (Read error: 104 (Connection reset by peer))
03:26:06 Join Rob2222 [0] (
03:26:40 Join Darksair [0] (n=user@
03:30:27 Join alexbobp [0] (n=alex@
03:34:34 Quit Rob2222 (Read error: 104 (Connection reset by peer))
03:35:11LambdaCalculus37Modifying firmware/target/arm/system-pp502x.c and commenting out line 320 worked.
03:45:07 Join tsfriendship [0] (n=tsfriend@
03:46:56 Quit LambdaCalculus37 ("Ka-chunka")
03:49:12 Quit Darksair ("Use the Force, Luke!")
03:53:26 Join n17ikh|Lappy [0] (
03:56:10 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
04:01:55 Join VirtualDi [0] (
04:01:58 Nick VirtualDi is now known as Virtualdistance (
04:02:33 Quit qweru ("moo")
04:02:44Virtualdistancehey i heard sansa v2 has sound now
04:03:18 Join blkhawk- [0] (
04:11:03 Quit mc2739 ()
04:12:49 Quit jhulst (Remote closed the connection)
04:19:29 Quit blkhawk (Read error: 110 (Connection timed out))
04:20:16 Nick blkhawk- is now known as blkhawk (
04:23:37 Quit XavierGr ()
04:46:59***Saving seen data "./dancer.seen"
04:53:17 Quit perrikwp (" ajax IRC Client")
04:53:29 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
04:53:57 Part safetydan
04:56:47 Join webguest78 [0] (n=47e8e972@gateway/web/cgi-irc/
04:57:17 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
04:57:22 Quit n17ikh|Lappy ()
04:57:51 Join massiveH [0] (
05:01:12 Join blkhawk- [0] (
05:07:03 Quit miepchen^schla (Read error: 110 (Connection timed out))
05:13:17 Quit TMM (Connection timed out)
05:16:48 Quit tarbo (Remote closed the connection)
05:17:37 Join tarbo [0] (n=me@unaffiliated/tarbo)
05:18:17 Join TMM [0] (
05:18:47 Quit blkhawk (Read error: 110 (Connection timed out))
05:19:10 Nick blkhawk- is now known as blkhawk (
05:34:33 Quit Horscht ("IRC is just multiplayer notepad")
05:38:06 Join Rob2222 [0] (
05:45:03 Quit webguest78 ("CGI:IRC (EOF)")
05:53:14 Quit planetbeing (Remote closed the connection)
05:54:09 Quit nplus (Read error: 104 (Connection reset by peer))
06:00:41 Quit dberg918 ("Ex-Chat")
06:09:15 Join Chaosmagi [0] (
06:10:00Chaosmagi Do u feel like your life is stuck in a rut, just going around in circles. Do u feel Spirituality left out then all u have to do is !!!!TAKE BACK REALITY!!!!
06:10:03 Quit Rob2222 (Success)
06:10:22 Part Chaosmagi
06:10:24 Quit pixelma2 (Nick collision from services.)
06:10:33 Join pixelma2_ [0] (n=marianne@rockbox/staff/pixelma)
06:10:34Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
06:10:35 Nick pixelma2_ is now known as pixelma2 (n=marianne@rockbox/staff/pixelma)
06:10:41 Quit amiconn (Nick collision from services.)
06:10:45scorche+b *!*
06:10:46 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
06:10:52Mode"#rockbox +b *!* " by scorche (i=Blah@rockbox/administrator/scorche)
06:14:53 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:19:02 Join Shaid [0] (
06:20:05 Quit Shaid (Client Quit)
06:23:34 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
06:31:25Virtualdistancehey sounds workingon sansa v2 e series right?
06:32:16LloreanVirtualdistance: Current status is in the forum thread and on the associated wiki pages.
06:32:35Virtualdistanceso no build yet?
06:32:45LloreanIf it was supported, it'd be listed on the front page as supported...
06:32:56Virtualdistanceis there an unsupported one?
06:33:07LloreanYou can compile a development one if you're planning on working on it, sure.
06:34:16Virtualdistancei guess i just need one compiled bleeding age "in development build"
06:34:51 Quit JdGordon (Remote closed the connection)
06:34:56Virtualdistancecause i rather have a bleeding age one than OF :)
06:35:46LloreanIf you can't compile it for yourself, you really shouldn't be using it yet.
06:36:23 Join JdGordon [0] (
06:44:01LloreanHave some patience. When it's reliable enough for use from day to day, it'll move into "supported"
06:47:03***Saving seen data "./dancer.seen"
06:54:29 Join AndyIL [0] (i=AndyI@
06:55:23 Quit AndyI (Read error: 60 (Operation timed out))
07:11:29 Quit perrikwp (" ajax IRC Client")
07:11:49 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
07:26:32 Quit tsfriendship (Read error: 104 (Connection reset by peer))
07:33:35 Join tsfriendship [0] (n=tsfriend@
07:34:38 Quit safetydan ("Leaving.")
07:41:40 Join ball [0] (
07:42:39ballmy iPod somehow booted into the Apple firmware instead of Rockbox. Is there a "key combination" that will fix it, or do I need to go through the RockBox install procedure again?
07:43:02scorchejust a reboot should do it
07:43:33 Join kugel [0] (n=chatzill@unaffiliated/kugel)
07:43:33ballscorche: how do I do that?
07:44:00 Quit kugel (Client Quit)
07:44:12scorchehold menu and select...this process is detailed many times throughout the internet and the would be prudent to do a bit of searching..
07:46:19ballokay, I'll look for TFM to R
07:49:18BHSPitMonkeyTFM is a very obvious part of the site's navigation bar...
07:49:46ballThere's a site? :-)
07:50:08BHSPitMonkeydon't make me facepalm
07:51:11*ball grins
07:51:22ballI know I looked at the site when I installed Rockbox
07:51:54ballSadly college means I don't get to read everything that I would like to.
07:53:35ballI'm going to bed now. I'm too tired to think.
07:54:57 Quit ball ("leaving")
07:59:09 Quit Seed ("cu, Andre")
07:59:30 Join Outerlimit [0] (
08:16:56 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
08:17:23 Quit BigBambi (Read error: 104 (Connection reset by peer))
08:19:04 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
08:20:49 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
08:25:39 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
08:28:55 Join ender` [0] (
08:29:28 Join hopit [0] (
08:33:36pondlifejhMikeS: Any chance that your PCM muting mod will have fixed keyclick on those devices where it was broken? Or shall I carry on assuming that's a PP problem...?
08:40:42 Quit amiconn (" ibb")
08:47:06***Saving seen data "./dancer.seen"
08:47:40 Join bertrik [0] (
08:51:45 Quit scorche (Read error: 104 (Connection reset by peer))
08:52:51 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
08:57:46 Quit bertrik ("Leaving")
08:58:40JdGordonjhMikeS: there is a bug report which is possibly caused by your pcm muting fix (incase you didnt see it)
09:02:11 Quit GodEater (" ajax IRC Client")
09:02:17 Join GodEater [0] (i=c2cbc962@gateway/web/ajax/
09:09:51 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
09:19:27 Quit scorche (Read error: 104 (Connection reset by peer))
09:20:58 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
09:22:34 Join petur [50] (n=petur@rockbox/developer/petur)
09:23:21 Join einhirn [0] (
09:27:09 Quit kachna (Read error: 60 (Operation timed out))
09:33:35 Join lasser [0] (
09:44:00 Join Thundercloud [0] (
09:45:13 Quit JdGordon (Read error: 104 (Connection reset by peer))
09:50:17 Join JdGordon [0] (
10:05:22 Join TheSphinX^ [0] (
10:07:15 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
10:20:51*JdGordon thinks that some time after 3.1 we need to have a discussion about how we eventually want to allow themeing in the rest of the ui, at the very least how we want to allow custom list viewport
10:22:02*Zagor has toyed around with a "transitions" idea
10:23:33 Join kontroll [0] (
10:24:08Zagornot really in response to your statement (which I agree with). just that it's a simple form of eye candy that I think many would like. and it could even be made pluggable.
10:24:23pondlifeJdGordon: Would you think it better if the wake-up alarm used the same setting code as the clock - allowing setting of an exact time (and probably reducing binsize) ?
10:24:34Zagorlike the left/right slide on ipods. or fade in/out, or crumble or whatever anyone wants to do.
10:24:40kontrollQuick question; what's the battery life like for rockbox on the iPod Video (60GB) compared to the default firmware?
10:25:06pondlifeI'd like to see the entire UI reworked to use a central state machine - so each screen is called from one place only and returns a "next screen" value...
10:25:29Zagorpondlife: whoa
10:25:35JdGordonpondlife: yes, and iirc there is a patch which does that, although its probaly out of sync
10:25:45B4gderpondlife: I'd like that too
10:25:54JdGordonpondlife: well the root menu mostly did that
10:25:56pondlifeI don't mean each and every screen (e.g. submenus would be part of the menu screen), but consistency could be improved
10:26:03B4gderkontroll: pretty good I think, there's a wiki page listing runtimes somewhere...
10:26:09pondlifeJdGordon: Yes, but we need to go all the way ;)
10:26:11Zagorit will be one heck of a state machine
10:26:25kontrollB4gder: Will check, I didn't see any mention of it in the FAQs.
10:26:43pondlifeProbably not as bad as you'd think... we have some messy code around that should vanish...
10:27:08JdGordonI was thinking more about how the bars and list positions should be allowed to be set, and if we want to really go nuts with the WPS and either allow it to have 1 massive one which handles everything, or mutliple wps', one for each bit of the ui.. or what
10:27:34B4gderkontroll: but all the numbers seem to be very old...
10:27:53kontrollHow the hell did you find that so fast? :X
10:28:15B4gderyou should try this new cool thing called google! ;-P
10:28:23kontrollBut you're right. No 3.0 numbers, right?
10:28:35kontrollBah, google. I relied on the actual page structure.
10:28:46kontrollGuess I should have known to trust my googlefu.
10:29:06B4gder" [keyword]"
10:29:51kontrollAh. I don't normally rely on any advanced search features.
10:34:10 Quit Thundercloud (Read error: 110 (Connection timed out))
10:34:11B4gderI'll sign you up for my google class ;-P
10:36:22kontroll"Very stable. Battery life has been improved, but there is still room for improvement." from the IpodPort page anyway. Not sure when the table was updated, but if it's for 3.0, I'll be happy. Battery life was my one gripe with Rockbox.
10:36:52B4gderI think battery life has improved even after 3.0, especially for mp3 playback
10:37:48kontroll*shrug* It's been a while since I used lossy formats. ;)
10:37:55kontroll<3 me some FLAC.
10:38:20B4gderflac playback should be decent too
10:38:29kontrollI hope so. :)
10:45:54 Quit bmbl (Read error: 104 (Connection reset by peer))
10:47:10***Saving seen data "./dancer.seen"
10:51:57 Part pondlife
10:58:34 Quit nuonguy ("This computer has gone to sleep")
10:59:20 Join kachna [0] (
11:27:54 Join kugel [0] (n=chatzill@unaffiliated/kugel)
11:28:21*kugel slaps JdGordon
11:28:33kugelI was right :)
11:28:38*JdGordon punches kugel in the face and then stomps on his head while he is down
11:33:43kugelJdGordon: callback is a good idea. But it's not acknowledged if you go into a menu by going back (except for the root menu apparently). That's what my tests showed
11:34:18JdGordonno, not a menu callback... chanigng the theme should send a event
11:34:58kugelI tried sending GUI_THEME_HAS_CHANGED, but it apparently didn't work
11:35:16kugelin the callback function though, so that might be the reason it didn't work
11:35:28JdGordonbecause nothing is registered for it...
11:35:39JdGordonthat entire system needs doing
11:39:29kugelwhat about staying in the theme folder after changing a theme? if you change a font you can right away chose another font if you noticed it doesn't fit your current mood or whatever. I'd like to have that for themes too (and I know how to do it, so it's just a question of "would it be committed?")
11:39:50 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
11:40:12 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
11:41:02 Quit B4gder (Read error: 104 (Connection reset by peer))
11:41:09 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
11:49:49 Quit einhirn (Read error: 104 (Connection reset by peer))
11:52:32 Join n1s [0] (n=nils@rockbox/developer/n1s)
11:53:07 Nick fxb__ is now known as fxb (
12:16:51kugelJdGordon: have you seen my latest customlist version?
12:19:29kugelI think it's a good version now
12:20:18 Join funman [0] (
12:28:04n1sfunman: does the flash buffering patch take care of voice?
12:28:28funmanno idea
12:29:17n1snot important at this stage but just as a note, a on-the-fly voiceclip loading scheme is used on the ondio which could perhaps be adapted
12:31:21 Join moos [0] (
12:34:33 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:36:27n1sbtw, do you still have the large pcm buffer? maybe that can be shrunk to free some ram?
12:37:49funmanyes, but how this free ram would be used ?
12:38:54 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
12:39:24n1smaybe you need a static buffer to load a voice clip into for example?
12:40:32n1show much ram does the core need on the clip?
12:41:31funmanRAM usage: 879440
12:41:58funmanincluding 192kB for buffering_flash
12:42:50n1sif plugin buffer is 512kB and the pcm buffer is something like that in size you are already near the limit
12:43:23funmanI reduced plugin buffer to 384kB and pcm buffer is just taking whatever available RAM remains
12:44:03n1spcm buffer != audio buffer though, pcm is the one for decoded audio
12:44:31funmanand pcm buffer is a part of audio buffer?
12:46:20n1si'm not exactly sure how it's allocated, see apps/pcmbuf.c
12:46:49n1sbut it varies in size with crossfade so maybe you will have to drop or limit that
12:47:14***Saving seen data "./dancer.seen"
12:49:30funmanyes it's in audiobuffer
12:50:33n1smaybe that has something to do with the weird crashes, decoded audio overwritign stuff?
12:50:55funman_very_ probably, i'll try adding some sanity checks
12:53:17 Join robin0800 [0] (
12:56:02kugelJdGordon: I found a way to phase out menu_vp
13:03:01kugelmaybe not
13:12:16funmanI have ~280kB of buffer for encoded data, perhaps buffering.c assumes more (for example when setting the watermark at 512kB)
13:16:04 Join nplus [0] (n=nplus@141.25.Globcom.Net)
13:17:17funmanI see a failed assumption in bufgetdata()
13:17:21JdGordonkugel: yes, by having a viewport manager of sort which will give you the "fullscreen" viewport if one isnt passed in
13:22:10kugelJdGordon: no, by letting vps to lists.parent after initializing the list. The list code takes care of NULL pointers
13:22:55kugelso no matter if the list was initialized with NULL or a parent, vps points to a valid viewport after
13:23:26 Quit robin0800 (Remote closed the connection)
13:24:07kugelI get segfaults after exiting a plugin (in the sim), even though the adresses are fine
13:24:46 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
13:27:54 Quit linuxstb (Read error: 110 (Connection timed out))
13:30:45funmanbuffering.c bufgetdata() says "prep_bufdata ensures copy_n <= GUARD_BUFSIZE"
13:31:06funmanbut this is only true if the requested size is > GUARD_BUFSIZE
13:32:19funmanI suppose that wouldn't be a problem if we wrap around the buffer at buffer_len - GUARD_BUFSIZE but I can't tell if it's true
13:34:42n1sfunman: i think Nico_P is the one that knows that stuff best
13:40:17jhMikeSJdGordon: 9602?
13:44:27JdGordonjhMikeS: yeah, mpegplayer crash if music wasnt stated
13:45:45jhMikeStechnically, the code shouldn't have changed for e200 since it never had the muting turned on
13:46:57 Join gregorovius [0] (
13:47:27funmanh->ridx = 2MB+
13:48:02kugelJdGordon: with that change we could really get rid of list_default_init_viewports
13:48:20kugelI don't understand why it segfaults though upon exiting a plugin
13:48:35funmanNico_P: why are ridx and available (in struct memory_handle) volatile?
13:48:47JdGordonno, i dont tihnk that change would let you get rid of that
13:49:16kugelJdGordon: I just removed them and the menus refresh fine upon theme change
13:49:55Nico_PJdGordon: funman, IIRC, so that gcc treats it correcty as it may be modified concurrently
13:50:00Nico_Pmaybe it's overkill though
13:50:10kugelthat's because vps really points to the list.parent, and not to the parent passed to the menu or a seperate menu_vp
13:50:19JdGordoncrosstalk :D
13:50:26Nico_Psorry JdGordon, I didn't mean to highlight you
13:50:44funmanNico_P: I try to find the problems in buffering.c but the code is quite complex :/
13:51:21Nico_Pfunman: I don't really have time to answer your questions rght now as I have to go to classes
13:51:24funmanNico_P: why would it be modified concurrently ?
13:51:28Nico_Pbut this evening I should have some tim
13:51:43funmanhave fun ^^
13:52:02Nico_Psome funcs like bufopen and bufalloc are called on threads other than the buffering thread
13:52:57JdGordonkugel: ah yes... you're right... and its even worth doing... lemme know when a patch is ready
13:53:21JdGordonbtw, it works because the menu list is re-init (which default_blaa is going anyway
13:53:23Nico_Phmm on second thought I'm not sure that's actually relevant... I'd need to give another look at the code. it's been a long time
13:53:31JdGordonbut it means 1 less static viewport struck wasting ram
13:54:02 Join AlwieAlwie_ [0] (
13:54:19AlwieAlwie_was ist das?
13:55:10n1sAlwieAlwie_: English please
13:55:22AlwieAlwie_Where I am? xD
13:55:29JdGordonabout to be kicked
13:55:40funmanAlwieAlwie_: just have a look at the topic
13:56:00AlwieAlwie_What is rockbox?
13:56:00gregoroviushow well does rockbox work in a 5g ipod? are the controls playable?
13:56:27B4gdergregorovius: it works fine and stable on the 5g
13:56:43gregoroviussorry, mistype, rockboY
13:56:46 Quit AlwieAlwie_ ()
13:56:56B4gderI don't know
13:57:03gregoroviusI was just browsing the plugins page
13:57:03 Join tyfoo [0] (
13:57:14gregoroviuswondering if games are playable
13:57:21JdGordonkugel: almost willing to bet that the buttonbar will cause problems with that change though
13:57:23kugelJdGordon: sure, when I find out why plugins crash. Also, do_menu unnecessarily clears the display/viewport I think
13:57:28 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
13:57:42kugel(list drawing does that too)
13:58:30JdGordonits done because screens dont clear themselves before exiting...
13:58:39JdGordonso you have scrollingin lines from the old screen
13:59:16funmaniCatcher.128x64x1.160x128x2.wps < does this naming reflects screen sizes/depth for remote and main lcds?
14:01:25funmanin wps/ I see files are named "theme.width.height.depth.wps"
14:02:04funmanand 2 icatcher files are named with 2 different "width.height.depth"
14:02:55funmani'm trying to build a different wps for the Clip
14:02:58pondlifeJdGordon: My rough state machine/screen transition idea would take care of clearing stuff...
14:03:06 Join LambdaCalculus37 [0] (i=44a04303@gateway/web/ajax/
14:03:47kugelJdGordon: Got it! not handling bars caused the crash
14:03:50n1sfunman: i thik the first screen dimensions are for a remote corresponding to the wps use on the main display with the second dimensions
14:03:55*pondlife wonders if he's the only one who'd like consistent braces throughout Rockbox...
14:04:23JdGordonfunman: oh right... n1s is probably right.. i guess because we can only have 1 userfont for both screens
14:04:53JdGordonpondlife: how? the central logic would clear the screen automatically?
14:05:22pondlifeEspecially if a scroll/crossfade transition was configured :)
14:05:57pondlifeWe'd have a single place to do it, no need to rely on the screen itself.
14:06:14pondlifeThis is a *very* rough idea, of course!
14:06:34JdGordonroot_menu.c does what you want now already....
14:06:43JdGordonexcept the screen clearing
14:07:42pondlifeYes, but is it used for all transitions? Some screens (e.g. context menus) are bool functions - I'd think they could be int functions returning a "next screen" (of which "previous" is a valid option)
14:07:52 Quit TheSphinX^ ("XChat@Linux")
14:08:03pondlifeSomething monstrous like that, anyway
14:08:28pondlifeSorry, I meant enum, not int...
14:08:33JdGordonwell.... context menus are directly linked to their host screen, and don't really need to go through the central logic
14:09:08*JdGordon wants multifont :<
14:09:17n1spondlife: I'd like consistent braces but as it stands we have pretty much every style used somewhere however the opening brace of a function is a strict rule in CONTRIBUTING
14:09:51B4gderconsistent braces would open for an "interesting" talk on what specific approach to use
14:10:09pondlifen1s: I've always worked to a policy of "maintain the existing style", but I'd prefer to use separate lines throughout.
14:10:22pondlifei.e. no if (x) {
14:10:36pondlifeIt's a personal preference, of course...
14:10:46n1spondlife: i prefer that style too but i guess it is a matter of taste
14:10:49*JdGordon thinks brace positions should be lower on everyones todo list...
14:10:51funmanperhaps braces style should be a setting?
14:11:14funmannote using a newline increases binsize by 1
14:11:38JdGordon1? my tests showed eleventy billion
14:11:38pondlifefunman: srcsize, shurely?
14:12:36n1swe should cut down on whitespace and comments to save srcsize!
14:12:45pondlifeYes, and use tabs
14:13:09JdGordonand get rid of #defines to speed up compiling
14:13:11jhMikeSn1s: and all those newlines while we're at it. C really doesn't need them.
14:13:11pondlifeMore seriously, I like the if (1==x) style of testing
14:13:43 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
14:13:43JdGordonthat one makes sense
14:13:47JdGordonalthough is annoying
14:13:58B4gderI like that one too
14:14:01n1sjhMikeS: yeah, why waste lines? :)
14:14:13B4gderC does need them
14:14:18B4gderat least in C89
14:14:22jhMikeSpondlife: it's my personal preference so I don't mess up if (x = 1)
14:14:23pondlifeIt becomes less annoying if used everywhere.
14:14:26B4gderit only supports 512 byte lines
14:14:40pondlifejhMikeS: Yes, I learnt that from you, now I use it in my day job
14:14:48n1sB4gder: we need to extend gcc then!
14:15:02jhMikeSpondlife: you did? I thought I didn't use that here. ?!?
14:15:04B4gdern1s: they fixed that later on so modern C versions can have any lengths I believe
14:15:17pondlifejhMikeS: Off someone here, I thought you explained it to me...
14:15:36pondlifeB4gder: Nah, we'll never get the punched-card machine port completed.
14:15:52JdGordonkugel: patch? or tomorow and ill go to bed?
14:15:55jhMikeSpondlife: in an alternate worldline before they start collapsing it could be
14:16:30pondlifeMaybe, I guess the IRC logs contain all possible truths.
14:17:05pondlifeAnyway, sorry for mentioning braces - let's fix bugs instead...
14:17:06 Quit perrikwp (" ajax IRC Client")
14:17:52kugelJdGordon: wait a second. some plugins still crash, but that might be due to my other changes
14:18:00*pondlife wants to commit the H300 RTC patch, but can't remember if that's officially blocked by petur's hard disk.
14:18:01jhMikeScode formatting is so much rewarding than proper behavior :)
14:18:02n1sLambdaCalculus37: regarding your ipd with ata -1 error, it would be useful if you worked out which bit(s) that actually cause that problem instead of simply removing the whole line. If that is figured out a fix could be put in svn, see FS #9342
14:18:07 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
14:18:46n1spondlife: the alarm wake up thing?
14:18:47JdGordonpondlife: what was the problem there? it wouldnt read the disk or something yeah?
14:19:00pondlifen1s: Yes, it works great for me, and has done for months.
14:19:18n1spondlife: problem is that it doesn't work with the official bootloader
14:19:28pondlifeUnfortunately it needs a new bootloader, but petur's hard disk won't boot with an SVN bootloader.
14:19:50JdGordonpondlife: but it can go back to the OF?
14:19:55*n1s hopes LinusN will release a new h300 bootloader some day...
14:20:03 Quit massiveH ("Leaving")
14:20:05LambdaCalculus37n1s: I just went with one fix that was listed for the time being just to get it up and running. I'll figure out what caused it later (could be the card I used, though).
14:20:14pondlifeSo, the patch is not committed and I have to add it manually every time, because we don't want RTC options appearing in the menus until the new bootloader is released.
14:20:25pondlifeHowever, I fear it will be years until this progresses.
14:20:32kugelJdGordon: works! no crashing anymore.
14:20:44 Quit funman ("leaving")
14:20:45kugeluntested on target though, lets rather wait until tomorrow. no need to rush
14:20:51LambdaCalculus37n1s: It's a kludgy fix; I admit it.
14:20:52n1sLambdaCalculus37: others have the same problem so a proper fix would be nice
14:20:58pondlifen1s: I don't think LinusN can repro the problem...
14:21:09pondlife...only petur can, IIRC
14:21:22JdGordonpondlife: as long as it can go back to the OF we should commit it, nag LinusN to release a new bootloader with it, then say if you have problems then use an old build and ignore the alarm options
14:21:36JdGordonkugel: can i see the patch anyway?
14:21:46pondlifeJdGordon: With the v5 bootloader it fails to boot up, I believe
14:21:56n1spondlife: but his disk works with the official bootloader?
14:22:01pondlifeYes, with v5
14:22:26JdGordonkugel: /me already doesnt like the first hunk :D
14:22:36pondlifeIt's an 8i0GB Toshiba one (MK8007GAH) - same as I have :/ Mine works fine.
14:22:56pondlifeRTC is a very useful feature...
14:23:04kugellist code calls viewport_set_defaults, no need for the menu to do it once more
14:23:08peturpondlife: LinusN knows what revision it starts going wrong iirc
14:23:17 Join nicfer [0] (n=usuario@
14:23:20kugelthe bar fixing stays
14:23:20JdGordonbecause if you're given a parent you shouldnt be modifiying it.. although i see that svn is doing it also
14:23:36peturpondlife: do commit whatever changes you need, I'll stick with the older bootloader
14:23:59peturbut I wasn't the only one with the issue iirc
14:24:01pondlifepetur: Yes, but until we work out what's wrong, we can't release a new bootloader.
14:24:13pondlifei.e. 50% of H300s might not boot with SVN.
14:24:17kugelJdGordon: why not? it's just about bar fixing and reversible
14:24:27peturit would be far less than 50%
14:24:40kugelmodifying the parent directly allows to save inits
14:24:55pondlifeI think the RTC menu options need to appear if you have an RTC bootloader...
14:25:11kugelsee how I commented out the init_default_* as they're not needed anymore
14:25:12pondlifeNot sure how to pass that info across though.
14:25:48pondlifeJdGordon: Quick question about the time/date screen - how can you tell if the wake-up alarm is set or not?
14:25:59*petur is off again to give work some attention
14:26:07JdGordonpondlife: you probably cant...
14:26:14kugelJdGordon: and yea, svn modifies parents too, not in the menu code, but in the list code
14:26:43pondlifeI should probably work on that to display the current setting on entry.
14:27:00pondlife(ie. a time, or "None")
14:27:21JdGordonit would be nice to stick a alarm icon on that clock to show if one is set
14:27:52pondlifeI'd like to be able to set any time too - like you can with most alarms
14:27:58pondlifeNot just 5 minute increments
14:28:02kugelI suppose we could replace the icon before the list entry with a "alarming one", or yea, put one next to the clock
14:28:22*n1s likes the statusbar idea
14:28:44Zagormy phone has a simple bell icon to show alarm is active
14:28:45n1soh and the compulsory wps tag
14:29:07n1smine displays a bell and the alarm time on the "desktop"
14:29:07pondlifen1s: I'll be happy with a WPS tag and a themeable status bar ;)
14:29:26pondlifeNo need to worry about updating the hard-coded status bar..
14:29:38n1spondlife: yes, that would be nice
14:30:02pondlifeDid anything come of linuxstb's proof-of-concept there?
14:30:14JdGordononly ideas....
14:31:03pondlifeI'll stop veering from subject-to-subject now and look at the alarm setting screen.
14:31:19kugelbtw: I wonder what happened about this one patch, with which you can display all settings in the wps
14:31:21JdGordonthere is a patch to merge the clock and alarm sceen codes
14:31:33pondlifeJdGordon: That would be good
14:31:44pondlifeIs it still usable, with your recent work?
14:31:54*JdGordon doesnt know its number or if its still in sync
14:32:08kugelit's rotting on the tracker and noone ever talks about it. I think it's a nice patch which possibly saves a lot of seperate tags
14:32:18JdGordonmy work didnt remove code it probalby uses... only moved it
14:32:34JdGordonkugel: /me saw it and likes the idea... dunno about how it was done though
14:33:01pondlifepetur: I'd be interested to see what happens if you try the RTC patch ( with your existing (v5?) bootloader. i.e. just build Rockbox, not the bootloader, then set an alarm.
14:33:51*petur has some unknown bootloader version cooked up by LinusN
14:33:57pondlifeAh, ok.
14:34:06JdGordondoes anyone know how we would go about passing a version string from the bootloader to the main build?
14:34:16peturI could try the patch with svn - maybe it boots now ;)
14:34:40pondlifeI don't think SVN has been changed much.
14:34:45pondlifeIt works for me, though ;)
14:35:12pondlifeI believe you'd be going back to the OF soon after.
14:35:54kugelJdGordon: so like the patch or not? I also have some ideas to clean up list.c's a bit
14:36:22JdGordonI like the idea...
14:36:47JdGordonI really want to seperate the bars from the rest of the screen which would clean things up ALOT
14:37:02pondlifeJdGordon: Why aren't the bars in their own
14:37:17pondlife(Or, are they?)
14:37:22JdGordonthey are... sort of
14:37:42pondlifeThey should be, seems like an obvious way to isolate them a bit more.
14:37:44kugelthe main problem is imho that they don't update from alone
14:37:54JdGordonoh they will when I'm done with them
14:38:00*JdGordon will beat them into submission
14:38:09kugellist/menu/etc code always needs to issue updates
14:38:47pondlifeThe bar code should be independent of any screen (except for the basic positioning/hiding of the viewports), right?
14:39:06JdGordonshould be yes
14:39:26pondlifeheh, I take that as "*should* be" :)
14:39:37JdGordonwe could even put the statusbar at the bottom with some work!
14:39:47kugelvery little work
14:40:03pondlifeOf course, so the menu status bar can be in the same place as the cabbie WPS one...
14:41:33*pondlife returns to work..
14:41:37 Part pondlife
14:42:13kugelJdGordon: what exactly do you mean with "i like the idea...". are you expecting some drawbacks?
14:42:48*n1s wonders if Lear is going to complete and commit FS #8523, it seems like a worthwhile improvement
14:42:51JdGordonnot really... its all tied into how the bars are done... once thats fixed this will be done easily
14:43:51kugelyou will aways need to adjust the menu vps depending on the bars, no matter how seperate bar code is
14:44:30JdGordonno you wont :D
14:44:46JdGordonviewport_set_defaults() would give you the viewport that doesnt have bars in it
14:45:12JdGordonthe problem is there is still too much code using the NULL viewport
14:45:14kugeland then you overlap menus and bars? cool
14:45:30JdGordonno, the bars are outside the viewport
14:45:40JdGordonthe screen would be split into 2/3 areas
14:45:48kugelyou want to shift down menus if there's a statusbar at the top
14:45:52 Part B4gder
14:46:10 Quit kachna (Remote closed the connection)
14:46:10kugeldon't you?
14:46:25 Join kachna [0] (
14:46:31JdGordonyes, but it will be done for you
14:46:53kugelby what code? by the statusbar or list code? either way you need to adjust viewports
14:47:15***Saving seen data "./dancer.seen"
14:47:18JdGordonby a bit of code which isnt written yet
14:48:48 Part nicfer
14:50:50 Join balou [0] (
14:56:21 Quit kugel (Remote closed the connection)
14:56:42 Join kugel [0] (n=chatzill@unaffiliated/kugel)
14:59:48 Quit moos ("Rockbox rules the DAP world")
15:03:00 Quit kachna ("Konversation terminated!")
15:08:37 Quit kugel (Remote closed the connection)
15:09:30 Join Horschti [0] (
15:10:23 Quit Horscht (Nick collision from services.)
15:11:54 Quit BHSPitMonkey (Read error: 104 (Connection reset by peer))
15:20:44 Quit hopit ("Leaving")
15:22:28 Join robin0800 [0] (
15:25:18 Join Schmogel [0] (
15:26:39 Quit Bjoern-Erik ("leaving")
15:30:44 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
15:37:27 Quit tyfoo ("Carpe diem")
15:37:47 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
15:43:29 Join Rob2222 [0] (
15:51:40 Quit markun ("Lost terminal")
15:52:51 Join markun [50] (n=markun@rockbox/developer/markun)
15:57:26 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
16:11:40 Join culture [0] (
16:13:37 Quit krazykit` ("Connection reset by beer")
16:15:38 Join krazykit [0] (n=kkit@
16:24:05 Quit jhulst (Remote closed the connection)
16:30:32 Join funman [0] (
16:38:06 Quit robin0800 (Connection timed out)
16:47:08funmanTo have a different wps for models with the same screen resolution/depth, I think should be extended to check for the existence of "THEME.WIDTHxHEIGHTxDEPTH-MODEL.wps" before "THEME.WIDTHxHEIGHTxDEPTH.wps"
16:47:17***Saving seen data "./dancer.seen"
16:47:43 Join toffe82 [0] (
16:52:30 Join is4 [0] (
16:59:53 Join kachna [0] (
17:01:46 Quit Zagor ("Client exiting")
17:12:54 Join {phoenix} [0] (
17:14:18 Quit perrikwp (
17:14:18 Quit Llorean (
17:14:18 Quit SUSaiyan (
17:14:18 Quit tim__b (
17:14:18 Quit Zambezi (
17:14:18 Quit shodanX (
17:14:18 Quit preglow (
17:14:18 Quit DaCapn (
17:16:11NJoinperrikwp [0] (i=18ac0c41@gateway/web/ajax/
17:16:11 Join Llorean [0] (n=DarkkOne@rockbox/administrator/Llorean)
17:16:11NJoinSUSaiyan [0] (
17:16:11NJointim__b [0] (
17:16:11NJoinZambezi [0] (
17:16:11NJoinshodanX [0] (
17:16:11NJoinDaCapn [0] (
17:16:11NJoinpreglow [0] (
17:18:40 Join CaptainKewl [0] (i=2669ecc2@gateway/web/ajax/
17:18:44 Quit {phoenix} ("Konversation terminated!")
17:22:25 Join herrwaldo [0] (
17:23:52funmandoes rockbox ensure playback & recording can't happen simultaneously?
17:25:44 Part LinusN
17:25:45LambdaCalculus37AFAIK playback immediately stops when you enter recording mode.
17:26:09LambdaCalculus37FM radio continues to play in recording mode.
17:26:26funmanso it doesn't stop?
17:26:57funmanor FM radio continues to be active, but its output is redirected to recording, and not to headphones?
17:30:31LambdaCalculus37funman: Let me check again.
17:30:37*LambdaCalculus37 gets out his c250
17:32:15LambdaCalculus37funman: Entering FM radio and then recording mode still outputs to headphones.
17:32:29LambdaCalculus37So I guess you can listen to when you want recording to stop.
17:32:38funmanoh nice
17:33:56funmanfor sansa ams we only have 2 DMA channels and one is already reserved for SD
17:34:13funmanperhaps we should only use DMA for output and standard copy for input
17:35:52funmanon PP dma isn't used for SD though
17:36:19 Join jgarvey [0] (
17:36:41 Join kugel [0] (n=chatzill@unaffiliated/kugel)
17:37:06LambdaCalculus37funman: So we can still have headphone output for FM radio, even when recording?
17:37:22 Join miepchen^schlaf [0] (
17:38:14LambdaCalculus37Cool. :)
17:38:15kugelfunman: can't you use interrupts to put all into those 2 channels?
17:38:18funmanwell, when we'll have access to the radio ;)
17:38:25funmankugel: no
17:38:44*kugel thought so
17:38:58funmanperhaps if we transfer ridiculous amount of datas at each time but i don't think that's worth
17:39:24kugelwhy ridiculous amounts?
17:40:26funmansmaller than the FIFO size of I2SIN & I2SOUT
17:40:53funmanor at least the time to transfer must be smaller than the size of data in the FIFO
17:40:56kugeloh you mean ridiculously small
17:41:09funmanyes sorry :/
17:41:14kugelsounds doable to me
17:41:26funmansounds not worth to me ;)
17:41:26kugelbut maybe not worth it, yea
17:42:55kugelhow about: record to a buffer on the one channel, and in an alternating fashion write to disk or sent to headphones
17:43:04kugelfrom the buffer that is
17:43:23funmanI tend to prefer the reverse
17:43:39funmanand use more efficient DMA for more used playback
17:43:51kugelfunman: for normal music playback, sure
17:44:27kugelbut in radio you cannot buffer ahead. also the buffer data must be live at anytime, so you cannot interrupt recording
17:44:47funmanthe i2sin chip buffers for us
17:45:11kugelthat buffer is obviously do small, as you've said
17:45:19funmanjust like the i2sout buffer
17:45:32kugelI mean a buffer in ram
17:45:54funmanI don't understand what you mean sorry
17:46:45kugelI'm just having random ideas in my mind. better you ignore me at all
17:50:21 Quit EspeonEefi ("さよăȘら")
17:53:23 Quit miepchen^schlaf ()
17:56:04 Join unstable [0] (i=unstable@tor/regular/sid)
17:56:27unstableWhat is the biggest microsd card you can put in a Sansa e280 version1?
17:56:33 Join {phoenix} [0] (
17:56:53LambdaCalculus37Theoretically 32GB.
17:57:03LambdaCalculus37That's only with Rockbox, though.
17:58:04unstableLambdaCalculus37: Do they make 32GB microsd cards?
17:58:08 Join miepchen^schlaf [0] (
17:58:15unstableThe biggest I can find on newegg is 2GB
17:58:45funmanunstable: just google for "microsd 32GB"
17:59:22n1s16GB ones should be available at least
17:59:37n1sthey might also be called mirosdhc
18:00:40unstableI see, microsdhc is what it is.
18:00:50unstableThe highest capacity newegg has is 8GB
18:01:06unstableFor pretty cheap too, some are $15, some are $30
18:01:16n1sunstable: note that those will only work while inside rockbox and not in the original firmware so you need a cardreader to put files on them easily
18:01:45n1s(or you could copy files to the internal storage and use rockbox to copy to the card but it's cumbersome)
18:01:51unstableright, I know it will only work in rockbox.
18:01:57unstableMy laptop has a built-in card reader.
18:02:06n1sgood :)
18:02:54funmanThe sansa ams has a spdif receiver, but I somehow doubt it is used
18:03:56n1sgood opportunity for a hw mod then :)
18:04:12unstablewow, $33 USD for 16 gigs of flash. Am I click on the wrong thing or what, will that work with the e280 v1?
18:04:19unstableI feel like that's way too inexpensive.
18:04:38n1sthat's not micro
18:06:18unstable$82, about the same price as the player
18:06:49unstableAnyone have one of these e280s? Will this card bulge out of the player?
18:06:57LambdaCalculus37unstable: <−− This is more of what you're looking for.
18:07:09LambdaCalculus37And it sits flush inside.
18:07:19 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:07:32 Quit petur ("work->home")
18:07:37unstableLambdaCalculus37: But that is only 8GB
18:10:10 Quit domonoky (Read error: 104 (Connection reset by peer))
18:10:15funmanhum I'm not sure how to handle concurrent access to the destination buffer pointer and size; between the i2sin interrupt handler and pcm_record_more()
18:10:33 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:10:50funmanwell I suppose that's why pcm_rec_*lock() are there after all
18:11:15 Quit linuxstb (Read error: 145 (Connection timed out))
18:12:27 Quit gregorovius ()
18:14:31 Join vitja [0] (n=vitja@
18:20:21 Join faemir [0] (
18:23:16*gevaerts just came back to his laptop to see settings_list.c scrolling by...
18:29:12 Join karashata [0] (n=karashat@
18:42:30 Quit perrikwp (" ajax IRC Client")
18:42:44 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
18:45:23 Join Seed [0] (
18:47:19***Saving seen data "./dancer.seen"
18:51:18 Join MethoS- [0] (
18:51:31 Join tyfoo [0] (
18:55:41 Join nuonguy [0] (
19:01:25 Join bertrik [0] (
19:03:53 Quit jhMikeS (Read error: 104 (Connection reset by peer))
19:05:04 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
19:06:30funmanI'm trying to record on sansa ams using the microphone but I get a I2SIN FIFO push error
19:07:49domonokycheck if you get a fifo overflow/ fifo full interrupt before this...
19:08:23funmanhm i'll just show the raw status
19:09:20 Quit tyfoo ("Carpe diem")
19:10:09domonokyprobably the fifo just overflows, as we got the same backwards at the beginning of i2sout support. (fifo empty, then fifo pull error)
19:11:25funmanyep it's full
19:12:27domonokyso your dma transfers dont work correctly.. :-)
19:12:42funmani don't want to use dma
19:12:53funmanso we could record and play at the same time
19:13:22domonokyah, you are reading it directly and interrupt driven ?
19:13:38funmanperhaps it will work at low samplerates
19:14:35domonokymaybe it would also work better if you use a fiq instead of a irq.. but i dont know whats missing to get them running.
19:15:31funmanjhMikeS also suggested we use vectored interrupts
19:16:57funmanbtw how does FS #9332 (buffering) work for you on m200v4 ?
19:17:43funmanit seems the interrupt happen even if i didn't enable it in VIC_INT_ENABLE
19:18:20domonokyfunman: seems to work ok.
19:19:25funmanperhaps pcm_lock() and pcm_unlock() fail as well
19:19:36bertrikI wonder about the HPCM in the codec, it is needed for the clip (i.e. not actively turned off), but apparently not for m200v4?
19:20:23funmani have a deadlock when I enter recording although :/
19:20:38bertrikdomonoky, does stereo on the m200v4 sound normal, or do you get a kind of echo-y effect with very little bass?
19:21:25domonokybertrik: it seems to be fine, but lets check which settings i use at moment.
19:24:30funmangit svn info reports revision 0 when the tree has been modified :/
19:29:50*domonoky thinks it sounds better with HPCM_off, but i dont really trust my ears :-)
19:30:04funmanfor the Clip the difference was very noticeable
19:32:07 Quit funman ("leaving")
19:32:16*domonoky should do the same hold button fixes on m200v4..
19:37:16jhMikeSfunman: you only have one DMA channel?
19:37:41 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:38:18*jhMikeS wonders who every though two was sufficient'
19:39:37kugelit apparently is sufficient. the of can do radio+recoding just fine
19:40:35kugeljhMikeS: you did the video overlay thingy in mpegplayer, right?
19:40:37jhMikeSbut if you're using one for ATA, then you can't have one for full-duplex audio
19:41:06*kugel nearly said wvs, but then realized he would get slapped by Llorean
19:41:16jhMikeSkugel: overlay?
19:41:32 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
19:41:52*jhMikeS named it WVS and remembers Llorean had some part in that
19:42:24kugelI'm officially not allowed to call it wvs here. So I say "video overlay"
19:42:43jhMikeSOSD has been used
19:43:09kugelhm, osd even better than wvs actually, lets try that :)
19:43:17 Join Rondom [0] (
19:43:55kugeljhMikeS: anyway. the anti-aliased font patch struggles with it on LCD_PORTRAIT targets, since you implemented a mono_bitmap_part rotated version
19:45:05kugeland since aaf use a alpha_bitmap_part function which isn't ported to rotated, it only displays pixel garbage.
19:45:35 Join J-23 [0] (
19:45:38kugelit's not too different from mono_bitmap_part, but still over my head. So I wanted to ask you if you could have a look at that
19:46:34 Join Thundercloud [0] (
19:47:56jhMikeSkugel: I had to do something since I wanted to use the users font for it and rotation isn't a core feature
19:48:18LloreanjhMikeS: I always thought it was OSD.
19:48:23LloreanI don't recall being involved in WVS
19:48:55*jhMikeS was just bs-ing
19:49:31kugeljhMikeS: I haven't said the rotated version was to be in the core. But to get aaf in a committable state again, the font rendering function has to be implemented in a rotated version (at least for the mpegplayer)
19:49:43*jhMikeS is 100% responsible for "WVS" from the brain drain of writing 10k+ lines of code for that plugin
19:49:53kugelreverting to sysfont in the mpegplayer, and only if a aaf is used, is surely not nice
19:50:35LloreanjhMikeS: I more or less asked kugel not to use "wvs" outside of when he's mentioning it function names that include it, until we start calling the screen the "wvs" in the docs, since it's not a term users or most others are going to recognize for the screen.
19:50:58 Quit blkhawk ("Caught sigterm, terminating...")
19:51:21jhMikeSkugel: I suppose it just has to swap things around so shouldn't be that hard. Personally, I think rotating should be a core feature even if functions need two versions.
19:52:00jhMikeSLlorean: Is there any official name yet? On-Screen Display is the common video term.
19:52:12LloreanOSD works fine for me.
19:52:26LloreanIt doesn't need an official "name" if there's an acceptable term like that anyway.
19:52:43LloreanIt's a common term used for the same thing in other software.
19:53:29LloreanI just wanted to avoid throwing around "made up" terms that we don't already define for our users in the docs except when strictly necessary.
19:53:33kugeljhMikeS: would you mind looking into it? For you, who made the mono version, it should be pretty straight forward, yes
19:54:42jhMikeSkugel: It's just about swapping the coordinates really. Which patch is it?
19:54:42 Quit faemir (Read error: 110 (Connection timed out))
19:55:04 Join faemir [0] (
19:57:38kugelI understand that it's mostly about swapping coordinates (I thought that too before I looked at it for the first time), but then it needed some more understanding for font stuff like glyphs
20:06:17jhMikeSunless I'm missing something, there's only the one patch, correct?
20:08:51 Join skipper [0] (
20:08:52kugeljhMikeS: the first version is pretty perfect already, it's just the mpegplayer problem
20:10:06kugelgreyscale is how much bits? 2 or 4?
20:10:42jhMikeSIt looks variable to me
20:11:14kugelI mean grey scale targets
20:11:20LloreanI think they're all 2bpp right now.
20:11:23kugelthe patch uses 4bit alpha blending
20:11:27jhMikeSHowever, something like the S could just render TTF vector fonts on the fly and be able scale them
20:11:48kugelit was supposed to be variable, but it turned out 4 is sufficient (2bit looked bad)
20:12:50 Quit TMM (Read error: 104 (Connection reset by peer))
20:13:48kugeljhMikeS: yea, if we had only beast
20:13:53kugel-like targets :)
20:15:53jhMikeSThere will be more I'm sure. I don't think hardware will get weaker as time passes and rockbox shouldn't fall behind the curve.
20:16:29kugelof course not
20:17:02kugelbut this patch will do for a) existing, not so powerful targets and b) until more beasts are supported
20:17:05 Join TMM [0] (
20:22:21jhMikeSwith the beast (when most grunt work is done), the IPU should be able to draw alpha-blended stuff.
20:23:32 Join blkhawk [0] (
20:24:12kugeljhMikeS: Ok, lets seperate the beast from the other targets and implement real men stuff for the beast only? :D
20:26:27jhMikeSkugel: I'm want to get to the fun stuff like that. mpegplayer is a perfect app for using that HW. It can do real overlays, rotation and YUV conversion.
20:28:33 Join TMM_ [0] (
20:28:38 Quit TMM_ (Read error: 104 (Connection reset by peer))
20:29:58kugeljhMikeS: I begin the believe it's stronger than my notebook
20:31:44 Join perrikwp|class [0] (i=98212b1c@gateway/web/ajax/
20:32:00 Join tvelocity[a] [0] (
20:32:27 Quit nuonguy ("This computer has gone to sleep")
20:34:47 Join Strife89 [0] (n=michael@
20:44:53 Quit tvelocity (Success)
20:45:32 Join fredddy [0] (
20:45:54 Join nicfer [0] (
20:47:21***Saving seen data "./dancer.seen"
20:49:44nicferHow much is the probability that a company decides to use rockbox as firmware for their player?
20:50:40krazykiti'd say closer to 3.1416%
20:50:49scorche|shbut seriously, i dont see how one can answer such a isnt a is a decision
20:50:52fredddykugel, I have trying your button code for the fuze. I got the power button working ( I think you got that work too ?). My thought is that gpio_ie has to be set to something else
20:52:05*domonoky thinks its NaN% :-)
20:52:30scorche|shdomonoky: ah...i am going with your answer now
20:52:34gevaertsdomonoky: surely it's 5+2i?
20:55:20jhMikeSe^(i*pi) + 1
20:59:54nicferThe biggest advantage would be having a device with free software from the box
21:00:31scorche|shsure...but that isnt up to us
21:02:28*jhMikeS will start a company when once the lottery check arrives (darn beaurocratic red tape)
21:04:37 Join webguest08 [0] (n=51429e85@gateway/web/cgi-irc/
21:04:50kugelfredddy: I don't think so
21:05:04kugelfredddy: try to get the power button working without gpiob_ie
21:05:09 Quit webguest08 (Client Quit)
21:05:25kugelfredddy: but thanks for looking into it. I'm nearly out of ideas
21:06:02nicferWell, wait until someone picks this
21:06:04fredddykugel, No I mean to change the OB define to something different
21:06:43kugellinuxstb: what's cheaper: display->update() or display_update_viewport() ? Given that you save the pre-calculations to be within the rect
21:07:19kugelfredddy: OB is just our button mask, it's the gpio pins which are used for the buttons
21:07:57kugeland you enable interrupts with IE, and only for the pins responsible for button reading
21:08:52fredddyhmm okay
21:09:20kugelfeel free to play around! I'm not stopping you from finding the truth!
21:11:50jhMikeSthe truth finds you, if you are ready to see. :p
21:11:57 Join tyfoo [0] (
21:12:04 Quit vitja ("Ex-Chat")
21:12:20 Join phfaty [0] (
21:12:27 Nick phfaty is now known as k0rnz (
21:12:36 Join Dornul [0] (
21:12:52kugeljhMikeS: Really? Then it must be the force which isn't with me
21:12:54 Quit k0rnz (Client Quit)
21:12:55 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
21:13:32DornulWill there be a Rockbox for the Sandisc Sansa View?
21:14:19Dornulthe current firmware is terrible
21:14:37scorche|shDornul: we dont make predictions, but a port is in progress
21:14:45*jhMikeS is just fixing red stuff
21:14:52gevaertsscorche|sh: For the view?
21:15:02*jhMikeS can get out the tarot cards and do a reading
21:15:07scorche|shgevaerts: sure...why not ;)
21:15:14Dornulsurely hope so
21:15:52gevaertsIsn't that one different from the AMS series?
21:16:01scorche|shDornul: that doesnt mean there will be a just means that some people have looked at it
21:16:06scorche|shgevaerts: it is
21:16:33DornulI'll keep my hopes up ;)
21:16:36gevaertsscorche|sh: I wouldn't call that "in progress" yet...
21:16:51*kugel was about to say the same
21:16:55scorche|shDornul: hopes wont make a port happen...people working on it would
21:17:24DornulI know but I can't do it. Don't have the skills :(
21:17:43bluebrotherget 'em ;-)
21:17:50 Join EspeonEefi [0] (i=eefi@STRATTON-TWO-EIGHTY.MIT.EDU)
21:17:55kugelNow I wonder why pf takes ages to prepare covers even if I don't have any album art
21:18:35kugeljust to return with error
21:19:24Dornulscorche|sh: thanks for the awnser
21:19:41 Quit Horschti (Remote closed the connection)
21:20:43 Quit perrikwp|class (" ajax IRC Client")
21:22:20 Quit Dornul ()
21:23:33 Quit Schmogel ("Miranda IM! Smaller, Faster, Easier.")
21:23:47 Join dany_21a_ [0] (
21:26:34 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
21:26:35 Quit Seed (Nick collision from services.)
21:26:41 Join Seed [0] (
21:27:32 Part nicfer
21:31:46 Join petur [50] (n=petur@rockbox/developer/petur)
21:32:45 Quit perrikwp (" ajax IRC Client")
21:32:59 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
21:38:48 Join funman [0] (
21:42:29 Join Xerion_ [0] (
21:42:49 Quit Strife89 ("Bye, all!")
21:43:39 Quit Xerion (Read error: 60 (Operation timed out))
21:43:39 Nick Xerion_ is now known as Xerion (
21:45:24 Join Genre9mp3 [0] (n=yngwiejo@rockbox/contributor/Genre9mp3)
21:46:33 Quit Seed (Read error: 110 (Connection timed out))
21:48:15 Join aarcane [0] (
21:48:29 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:51:04kugelthe cowond2 sim is a pain
21:51:36 Quit faemir (Remote closed the connection)
21:51:58 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
21:52:23 Join faemir [0] (
21:52:28jhMikeSthose stubs are now causing linker conflicts whereas they didn't previously
21:53:25jhMikeSthe bootloader ones anyway. sims and targets are ok now.
21:53:40jhMikeSeh, firmware, blah
21:55:05 Join stoffel [0] (
21:56:00 Quit faemir (Remote closed the connection)
21:57:01 Join faemir [0] (
21:58:23 Quit LambdaCalculus37 (" ajax IRC Client")
22:02:11 Quit {phoenix} (Remote closed the connection)
22:02:53 Quit puzzles ("byebye")
22:06:55 Join Zagor [242] (
22:11:32 Quit petur ("*plop*")
22:18:53 Quit lasser (Remote closed the connection)
22:27:34 Join massiveH [0] (
22:31:07 Quit karashata ("G'bye everyone!")
22:31:12linuxstbkugel: I don't understand your question about the cost of lcd_update() vs lcd_update_viewport() - they do different things, so you should use the appropriate one. But additions/subtractions are very cheap.
22:41:52 Quit fredddy ("Verlassend")
22:42:07 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
22:42:28 Part safetydan
22:44:09kugellinuxstb: I know they don't do exactly the same. But I'm wondering why Rockbox at different places does "set_viewport(vp);lcd_update_viewport();set_viewport(NULL()"
22:45:16linuxstbI don't know - hopefully I didn't write that...
22:45:29kugelI'd just do "lcd_update();set_viewport(vp)". I don't see a reason in just updating a part of the lcd, if (and hence my question) updating the whole lcd is probably cheaper in such a case
22:45:53kugels/probably/might be/
22:46:24linuxstbIf that part of the code is just responsible for drawing inside a specific viewport, IMO it's the Right Thing to call lcd_update_viewport(). But the code you quoted looks awkward.
22:46:25kugelI mean, lcd_update_rect contains several sanity checks, adds/subs and a loop in general
22:47:24***Saving seen data "./dancer.seen"
22:47:36kugelalso, I've yet to find a good explanation to intentionally only update parts of the screen
22:48:14kugelexcept the "it's proper to only update the part the code is responsible for"
22:48:18linuxstbA partial update is faster than a full update.
22:48:42kugelOk, that answers my question :)
22:50:57 Quit massiveH ("Leaving")
22:57:36 Quit domonoky (Read error: 104 (Connection reset by peer))
23:01:37 Join webguest93 [0] (n=4679efdb@gateway/web/cgi-irc/
23:02:44webguest93i just bought the 4th gen ipod nano...are there any open projects that will support this gen?
23:03:04webguest933.0 doesnt?
23:03:05webguest93that sucks
23:03:16webguest93is there any other open source like rockbox for the 4th gen?
23:03:20webguest93other then ilinux?
23:03:36krazykitno, no non-apple code runs on that device.
23:04:02webguest93the 4th gen doesnt use apple code?/
23:04:05webguest93what is it then?
23:04:14Bagderno non-apple
23:04:21Bagderthat means only apple code
23:04:37webguest93do like ipodlinux wont even run?
23:05:30 Quit stoffel ("leaving")
23:05:44Bagderhow many times do you want the answer served?
23:05:57krazykitwebguest93, let's make it very clear. you're stuck with what you have.
23:06:27webguest93ok cool
23:06:35webguest93thanks guys...sorry my retartedness
23:06:46Bagdersell the nano and get a better player
23:07:55 Quit aarcane ("Leaving")
23:11:45bluebrotherhmm. Is there a reason the h100 manual platform file is called "h1xx.tex" while for the h300 it's "h300.tex"?
23:12:24bluebrotherI'm tempted to change it to h100.tex ...
23:12:28funmanpixelma2: I had a look at the wps building but I see no special code for h100/h300 - I suppose the difference between the two is the size of the remote lcd
23:29:04 Quit tyfoo ("Carpe diem")
23:34:31Lloreanfunman: I thought the remote LCDs are the same size.
23:34:36LloreanThe main screen LCD is different
23:34:50Llorean220x176 on the h300 vs 160x128 on the h100 if memory serves.
23:34:52 Quit krazykit ("Connection reset by beer")
23:35:27 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
23:36:18 Join krazykit [0] (
23:38:14 Join ej0rge [0] (
23:38:14 Quit MethoS- (Remote closed the connection)
23:38:44 Join MethoS- [0] (
23:39:04advcomp2019ej0rge, hello.. you have a question?
23:40:28ej0rgeadvcomp2019: Thought I'd ask for wiki write access, so i can add a link to the newer, cheaper, smaller CF card to 50 pin ide adapter from dealextreme to the cfmodguide page
23:41:30advcomp2019what is your wiki name?
23:41:30ej0rgemaybe note my success in adding an RTC to my H120 to that page
23:41:34ej0rgeEric Jorgensen
23:43:49linuxstbej0rge: Do dealextreme ship worldwide? If not, it would be nice to mention that.
23:44:33 Quit Thundercloud (Remote closed the connection)
23:44:47advcomp2019ej0rge, done.. do not spam please too.. linuxstb, i think they do
23:44:57 Quit bertrik ("Leaving")
23:45:23ej0rgelinuxstb: I think they do. I see reviews in various languages on their site.
23:46:06ej0rgeadvcomp2019: wouldn't dream of spamming. I'm just a regular DX customer who's getting ready to install a CF card in a H320 that was brought back from the dead.
23:46:13 Quit krazykit (Remote closed the connection)
23:46:29funmanej0rge: don't worry, it's just a Turing test ;)
23:46:54ej0rgewithin a couple weeks i should be able to report whether the Topram 32gb CF cards that go for about $50 on ebay work w/ rockbox in irivers
23:47:19 Join krazykit [0] (
23:47:38Zagorej0rge: what was that about adding an rtc to your h120?
23:48:26ej0rgeZagor: You can add a Dallas RTC chip to an H120 very easily. There's a guide in the wiki. I did it purely for hack value, because the chip is available as a free sample from Maxim
23:48:53ej0rgeit was way easier to install the RTC than it was to install a schottky diode between the usb power pin and the DC inlet socket
23:49:27Zagorah, I hadn't seen that page
23:49:34linuxstbej0rge: To answer my own question, dealextreme are based in Hong Kong, and appear to ship anywhere...
23:52:20 Quit bmbl ("Woah!")
23:52:29saratogaBagder: any chance you could update the "Current Status" link on the front page to something like "Current Status and Other Targets"?
23:52:51 Join tyfoo [0] (
23:55:22Zagorsaratoga: "Current status on all targets"
23:56:03Zagor"for all targets" perhaps?
23:56:16 Quit krazykit (Remote closed the connection)
23:56:52Nico_PZagor: what about adding a link to MajorChanges?
23:57:26Zagorany suggestion where?
23:58:49 Part is4 ("Leaving")
23:58:57 Quit ender` (" PHP is a minor evil perpetrated and created by incompetent amateurs, whereas Perl is a great and insidious evil, perpetrate")

Previous day | Next day