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 2010-12-24

00:00:03 Quit benedikt93 (Quit: Bye ;))
00:00:03Clipperand i needed it just to remove one nasty virus
00:00:24kugelmaybe libusb?
00:00:48Clippertwo days later kaspersky released automatic uninstaller for that virus
00:02:25ClipperTorne: so ive used dependancy walker on rockboxutility.exe
00:05:27ClipperIESHIMS.DLL WER.DLL <−−- missing
00:06:11ClipperMPR.DLL <−−- something about wrong functions or missing
00:06:37 Quit wodz (Quit: Leaving)
00:08:01 Quit komputes (Quit: I haven't slept for ten days, because that would be too long.)
00:11:29soap <-Buschel, gevaerts, maraz
00:16:01Buschelsoap: added a short comment
00:16:44soapI'll check the upper 4 bits as soon as I figure out why said ipod is getting mounted read-only ;)
00:16:59***Saving seen data "./dancer.seen"
00:19:16BuschelI expect that w/o DMA 0x20000000 is not set at all
00:23:50soapok, where do I find that?
00:24:16Buscheloh, you'll need to add some code to the debug screen
00:31:00Buschelsoap: e.g. add a new lcd_putsf(...) to dbg_ports() in debug-pp.c
00:32:12Clipper30 is default rockbox contrast?
00:32:39 Quit n1s (Ping timeout: 240 seconds)
00:35:21soaphaha, Buschel, you overestimate me. I'll try.
00:36:38Buschelalways good to have new challenges ;)
00:37:11TorneClipper: the default contrast is different on different devices
00:37:21soapdon't tell me yet, let me test my build.
00:37:30Torneif you want to reset any setting to the default, you can long press select on the setting to get the context menu up, there's a reset to default option there
00:45:08moosrasher: ping?
00:48:10soapBuschel, 20150020 seem like a real result? ;)
00:48:38soap lcd_putsf(0, line++, "IDE0_CFG: %08lx", IDE0_CFG);
00:48:40soapis all I added
00:49:05gevaertsshould be good
00:49:07Buschelsounds good
00:49:31Buschelthis is with DMA, right?
00:50:06Buschelboosted it becomes 30150020 ?
00:51:21Buschelok, can you rebuild w/o DMA now?
00:52:08rashermoos: Yeah?
00:52:13 Quit utanapischti (Ping timeout: 260 seconds)
00:53:42ClipperIm trying to create radio station rpeset in my sansa clip+
00:53:49soapBuschel, working on modifying ata-pp5020.c
00:54:17ClipperProblem is, i dont know how to accept radio station name i assigned
00:55:21the_KyleClipper: hold select once the name is as you want it. It's a standard virtual keyboard there.
00:55:42*The_Seven wonders how to proceed with the ipod classic rockbox port
00:55:47the_KyleAlso, be sure to save your preset list.
00:55:49Buschelhow? comment the two lines as I proposed? to switch off DMA you just need to comment HAVE_ATA_DMA in ipodnano1g.h
00:56:15The_Seveni still have no idea what codec it uses, or how to access I2S
00:56:37The_Sevenshould i try porting things now, or continue digging up more hardware first?
00:57:29soapnow you've confused me Buschel. Should I be commenting out the lines in ata-pp5020.c or not?
00:57:59LloreanThe_Seven: sometimes just getting the UI running and interactive can get a lot of attention that can bring people in who might know or be able to discover details we can't, so I tend to lean toward "do what you can do, then investigate more"
00:58:26The_Seventhe question is if i can bring up the UI without an audio driver...
00:58:55gevaertsThe_Seven: I'd suspect a very minimal stub driver will be enough for that
00:59:19Buschelsoap: let me clarify -> we are talking of two things 1) disabling DMA (commenting HAVE_ATA_DMA in ipodnano1g.h) to measure IDE0_CFG w/o DMA, 2) test a possible solution (commenting out 2 lines in ata-pp5020.c)
00:59:21The_Seven...which needs to contain half a dozen defines at least :)
00:59:38soapahh, now I fully understand. That makes sense.
00:59:49soapsorry about the bone head.
01:00:08Buschelwell, just a bit of overlapped ideas ;)
01:02:59Buschelsoap: after all of this testing your nano should be warm enough to easily check the possible fix :)
01:03:00ClipperAnd how can i save calendar options?
01:05:02soap00158020 and 10158020
01:06:36The_Sevenhm, should i put the pagetable in iram?
01:08:08 Quit TheLemonMan (Quit: free(me))
01:08:39Buschelsoap: now enable DMA again and just comment line 124 in ata-pp5020.c (IDE0_CFG |= 0x20000000)
01:09:13moosrasher: Hi. I thought that genlang could handle the use of a string multiple times, but apparently I canot. I wanted to reuse one string instead of adding a new one. Do you know how hard it could be to fix this?
01:10:10soapjust 124? Not 1009?
01:10:30The_Sevenhm, actually it isn't very clever to let me do a (rather tedious, if done for the first time) port bringup instead of letting me do RE work :)
01:10:39Buschelsoap: yes, at a first shot just 124
01:10:50pixelmaClipper: do you really mean the calendar plugin or the time and date screen?
01:11:32pixelmaalso, maybe you'll find the answers in our manual, I don't own a Clip so can't tell you more anyway
01:12:11rashermoos: Not sure what you mean?
01:12:40soapI really should stop making everything and just make the binary
01:12:47Buschelsoap: ata_dma_set_mode() is only called once at startup. so, line 109 should not have relevant impact as this value will be set in the clock scaling with each call. but 0x20000000 will be kept until the next restart.
01:13:07Buschelsoap: yes, rockbox.ipod is needed. not all the rest
01:14:51saratogaThe_Seven: the Classic needs a page table?
01:15:29The_Seventhere is no other way (that i know of) to configure memory region cache attributes (basically prevent caching of MMIO)
01:16:12saratogadoes it actually use VM then, or can you just set one big page or something like that
01:16:17TorneYou can't *not* use the MMU on ARMs that have one unless you also don't use the cache
01:16:20moosrasher: I wanted to creat a new id to lang used in 2 situations. reffering to
01:16:25Clippercan i delete some of rockbox files? like games older?
01:16:37Tornesaratoga: you can just make a page directory mapping all of ram in 1MB sections
01:16:41Torneyou don't need second level page tables
01:16:47Tornebut you have to make at least that.
01:16:54The_Sevensaratoga: we'll probably configure it with 1MB granularity
01:17:01The_Seven(what torne just said)
01:17:05saratogamakes sense
01:17:27rashermoos: Well that must be possible. What errors are you getting?
01:17:51The_Sevenwith some tricks (and the main/codec iram split being on a 16K boundary) we might only waste 4K of iram for the page table
01:17:57rashermoos: Also, I don't know much at all about that part, so I'm probably not the most qualified to answer
01:18:09saratogahow much IRAM is there?
01:18:13soapPANIC Dir entry 6 in sector 4 is not free! A4 A2 ED 4C
01:18:14The_Sevenwe could also just put it into DRAM, but i have no idea how that will perform
01:18:19moosrasher: error: redeclaration of enumerator ‘LANG_FILESIZE’
01:18:21moosnote: previous definition of ‘LANG_FILESIZE’ was here
01:18:24saratogathats so much its not a big deal IMO
01:18:35saratoga128k is enough for most codecs
01:18:49 Quit pamaury (Remote host closed the connection)
01:18:56moosrasher: ok, thanks anyway, I'll try to catch Bagder or n1s if I can
01:18:59rashermoos: That's pretty much out of my league. No idea
01:19:22moosnp, thanks
01:19:34Buschelsoap: then change line 124 from |= 0x20000000 to |= 0x10000000
01:19:43*moos don't know why he thought that was possible
01:19:44TorneIt should be fine in DRAM, but the pagetable is 16KB
01:19:48Torneer, page directory
01:19:51Torneit's not like x86
01:19:57TorneARM maps 1KB per pagetable, not 4KB
01:20:05soapmaybe I got it too hot? I'm not using the same methodology to quickly heat that I was before.
01:20:05 Quit dfkt (Quit: -= SysReset 2.53=- Sic gorgiamus allos subjectatos nunc.)
01:20:28TorneIf it's just a few dozen section mappings then the TLB will hit most of the time and it won't actually have to read the memory often
01:20:32soapfonts are corrupted on reboot.
01:20:32Torneso, i don't think it needs to be in iram
01:21:01ClipperIs there any way to speed up audiobook read speed in rockbox?
01:21:23ClipperSansa Clip OF has such an option
01:21:33Clipperif its not in rockbox, i proudly suggest adding it
01:21:46LloreanClipper: A good idea is reading through the manual. You'll find a lot of basic questions answered there.
01:21:59ClipperLlorean: i am but i cant find that one
01:22:11The_SevenTorne: we have a 30-bit address bus here, so only the first 4K are actually used
01:22:33LloreanClipper: Are you actually reading through it, or are you searching for specific terms and hoping what you call it matches what we call it? :)
01:22:38TorneThe_Seven: ah
01:22:49 Quit timccc1 (Ping timeout: 245 seconds)
01:22:50The_Sevenso we could re-use the other 12K as codec IRAM
01:23:06TorneThe_Seven: well, you don't have to map it 1:1 anyway
01:23:14Buschelsoap: does changing 0x2... to 0x1... help?
01:23:25ClipperLlorean: i was searching for audiobook and i couldnt find it, but not i found timestretch option ;)
01:23:32The_Seven(yeah, that could cause very undesirable effects when memory >0x3fffffff is accessed
01:23:46Torneyou could split it
01:23:48The_Seventhat wouldn't be just wrapping any more but doing Weird Things(tm) instead)
01:23:58Torneit's ARMv6 right?
01:24:07Torneso there's two page directory base registers
01:24:12The_Sevenv5 i think, ARM9e
01:24:18Torneone for the bottom N, one for the top X-N
01:24:23Tornethen no
01:24:45TorneARMv6 you can split the address space and have two page directories for it (for user vs kernel mappings)
01:24:52Torneso you could put one in IRAM and one in DRAM :)
01:24:58Tornebut i don't think that exists on v5
01:25:10saratogaweee more arm9e
01:25:57soapBuschel, I'll try tomorrow. "Dinner" was just called up the stairs.
01:26:21The_Sevensaratoga: there's also ARM11 ahead (nano4g)
01:26:31Buschelsoap: ok, just keep me up-to-date :)
01:26:58soapWill do. Thanks for all your attention.
01:26:58 Quit factor (Read error: Connection reset by peer)
01:27:03 Join factor [0] (
01:27:07saratogahaving a flash based arm11 device would be fun to work with
01:27:33The_Sevenshould be similar to the beast (they apparently re-used an ipod touch SoC for that one)
01:27:39Buschelgood night
01:27:43 Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014])
01:27:52The_Sevenso it should be pretty damn powerful for that kind of device
01:28:43The_Sevenwee! more undefined references than my scrollback goes back!
01:28:59saratogaJdGordon: what needed to be fixed in the lang files for fs#11748?
01:29:45[AndrewR]lately when I build rockbox, it lags on GEN version.h, and if I run top I see "svnversion"
01:30:00[AndrewR]anyone know how I can speed that up?
01:30:02*The_Seven runs away from those dozens of problems and goes back to reversing the OF for now, trying to dig up the audio driver
01:30:19[AndrewR]it lags on svnversion for like 1-2 minutes sometimes
01:31:14 Join timccc [0] (~timccc@
01:31:55JdGordon|saratoga: I'll reply in the tracker
01:32:11 Quit DerPapst (Quit: Leaving.)
01:33:43 Nick The_Seven is now known as [7] (~TheSeven@rockbox/developer/TheSeven)
01:34:57kugel[7]: don't you need only a few stubs?
01:35:29[7]a few hundred stubs... and i'm running into problems with the ata driver architecture anyway
01:37:16 Quit mortalscan (Remote host closed the connection)
01:37:30 Join mortalscan [0] (~mortalsca@
01:39:58 Quit timccc (Ping timeout: 240 seconds)
01:40:00JdGordon|rasher: changing a string from "foo: none" to "foo: some text" will break lang and voice files righ?
01:40:20rasherI believe so
01:40:26kugeli doubt, have you maybe HWCAPS defined from c&p of another <target>.h file? these will pull in some stubs
01:40:29rasherNot 100% sure
01:40:32kugel[7]: ^
01:40:52saratogaI read that reply but have no idea what you're talking about
01:40:59saratogahow does the voice system work?
01:41:06saratogaor lang system i gues?
01:42:21JdGordon|my understanding is that .lng and .voice files have an index at the start which points to an offset where a given id is (the text or voice)
01:42:42JdGordon|if the LANG_ID is changed (one was added or order changes) the wrong id will get pointed to
01:43:00JdGordon|this is why the order in english.lang is important
01:43:55JdGordon|deprecating a string doesnt affect the order because an empty placeholder is put in and the old id isnt actually used in the code anymore
01:44:07JdGordon|but undeprecating a string probably has dangerous effects
01:44:16rasherJdGordon|: I'd ask B4gder about the none -> "something" thing
01:44:19kugelif it works this way I could rename an ID without problems, right? I'm asking because of moos' patch
01:44:33JdGordon|simple rename, yes
01:45:06rasherkugel: do it in all langs (langtool can do it for you iirc)
01:45:08rasheror sed, of course
01:45:11JdGordon|I'll be AFK by the time B4gder is back online, if you see the question please reply :)
01:45:20kugelbut from "rec: text" to "*: text" breaks, right?
01:45:48JdGordon|thats the question
01:45:48mooskugel: my problem is I want ti use the string in recording and in screen
01:47:52 Join timccc [0] (~timccc@
01:47:56moosthe same string
01:49:33JdGordon|the way to check if to look at the code that generates the .lng files and see how it handles none
01:49:50JdGordon|if it writes "" it is probably safe
01:50:11JdGordon|except people with outdated .lng and voice files will get no output for the line
01:50:15JdGordon|which is probaly just as bad
01:51:10 Join simonrvn [0] (
01:51:16kugelno output is better than wrong output
01:51:38JdGordon|not really
01:51:40JdGordon|both suck
01:52:08kugelthen you always need new voice files with every binary
01:52:18JdGordon|genlang is perl so im not looking at it to try to understand it :D
01:52:21 Quit timccc (Ping timeout: 264 seconds)
01:52:42kugelin which case we wouldn't have to worry about compatibility
01:52:47JdGordon|no, because langs almost never change in a way which makes old lang files not work
01:52:57JdGordon|that is by design
01:53:00kugelthere's no guarantee though
01:53:11 Quit moos (Ping timeout: 240 seconds)
01:53:22JdGordon|there is a fairly good implicit understanding that it wont happen
01:53:25 Join moos [0] (moos@rockbox/staff/moos)
01:53:26kugeland new strings without new langs/voice will give no output
01:53:35JdGordon|no, they give the english output
01:53:41JdGordon|again, by design
01:53:48kugelwell, langs yes, but not voice
01:54:27JdGordon|the point is we dont activly break lang and voice files
01:54:28kugelvoice can't have the strings that didn't exist at the time it was generated
01:54:41JdGordon|well obviously
01:54:51kugel-> no output
01:55:11JdGordon|VOICE ONLY
02:04:33 Quit Kitar|st (Ping timeout: 260 seconds)
02:07:31[7]good news: the ipod classic of contains the string "MeCCA"
02:09:46 Join Kitar|st [0] (
02:15:50 Quit antil33t ()
02:17:01***Saving seen data "./dancer.seen"
02:40:09DEBUGEOF from server (Network is unreachable) (snapshot: netstuff.c line 545)
02:40:09***No seen item changed, no save performed.
02:40:11***Started Dancer V4.16
02:40:11***Connected to on port 6667
02:40:11***Logfile for #rockbox started
02:40:11Mode"logbot :+i" by logbot
02:40:15***Server message 501: 'logbot :Unknown MODE flag'
02:40:15 Join logbot [0] (
02:40:15 Join the_Kyle [0] (~kyle@
02:40:15 Join dionoea_ [0] (
02:40:15 Join jepler- [0] (~jepler@emc/developer/pdpc.professional.jepler)
02:40:15 Join niekie_ [0] (~niek@CAcert/Assurer/niekie)
02:40:15 Join shai_ [0] (
02:40:15 Join JesusFreak316_ [0] (
02:40:15 Join Guinness` [0] (
02:40:15 Join Topy [0] (
02:40:15 Join Kitar|st [0] (
02:40:15 Join moos [0] (moos@rockbox/staff/moos)
02:40:15 Join simonrvn [0] (
02:40:15 Join mortalscan [0] (~mortalsca@
02:40:15 Join factor [0] (
02:40:15 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
02:40:15 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
02:40:15 Join Clipper [0] (53182eb2@gateway/web/freenode/ip.
02:40:15 Join Dreamxtreme [0] (~Dre@
02:40:15 Join Guest16638 [0] (
02:40:15 Join krazykit [0] (~krazykit@
02:40:15 Join Strife89 [0] (
02:40:15 Join designate72 [0] (
02:40:15 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
02:40:15 Join soap [0] (~soap@rockbox/staff/soap)
02:40:15 Join Horscht [0] (~Horschti@xbmc/user/horscht)
02:40:15 Join madalu [0] (~user@unaffiliated/madalu)
02:40:15 Join toffe82 [0] (
02:40:15 Join anewuser [0] (anewuser@unaffiliated/anewuser)
02:40:15 Join chattr [0] (
02:40:15 Join rasher [0] (~rasher@rockbox/developer/rasher)
02:40:15 Join ranmachan [0] (
02:40:15 Join MagusG [0] (
02:40:15 Join simabeis [0] (
02:40:15 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
02:40:15 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
02:40:15 Join pixelma [0] (quassel@rockbox/staff/pixelma)
02:40:15 Join amiconn [0] (quassel@rockbox/developer/amiconn)
02:40:15 Join Barahir [0] (
02:40:15 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
02:40:15 Join fyrestorm [0] (
02:40:15 Join mordocai [0] (~mordocai@
02:40:15 Join saratoga [0] (62e16704@gateway/web/freenode/ip.
02:40:15 Join [Saint] [0] (S_a_i_n_t@
02:40:15 Join YPSY [0] (
02:40:15 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
02:40:15 Join parafin [0] (
02:40:15 Join linuxguy3 [0] (
02:40:15 Join fred_2 [0] (
02:40:15 Join kkit|sh [0] (
02:40:15 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
02:40:15 Join jfc^2 [0] (
02:40:15 Join liar [0] (
02:40:15 Join rvvs89 [0] (rvvs89@pdpc/supporter/base/rvvs89)
02:40:15 Join scorche [0] (~scorche@rockbox/administrator/scorche)
02:40:15 Join evilnick_ [0] (
02:40:15 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
02:40:15 Join Rasi [0] (
02:40:15 Join mikroflops [0] (
02:40:15 Join merbanan [0] (
02:40:15 Join AlexP [0] (~alex@rockbox/staff/AlexP)
02:40:15 Join crwl [0] (
02:40:15 Join froggyman [0] (~seth@unaffiliated/froggyman)
02:40:15 Join knittl [0] (~knittl@unaffiliated/knittl)
02:40:15 Join sinthete1 [0] (
02:40:15 Join tmzt_ [0] (~tmzt@
02:40:15 Join ej0rge [0] (
02:40:15 Join avacore^ [0] (
02:40:15 Join Bushmills [0] (
02:40:15 Join FOAD [0] (~dok@
02:40:15 Join Zambezi [0] (Zulu@unaffiliated/zambezi)
02:40:15 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
02:40:15 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
02:40:15 Join n17ikh [0] (
02:40:15 Join B4gder [0] (~daniel@rockbox/developer/bagder)
02:40:15 Join guymann [0] (
02:40:15 Join thegeek [0] (
02:40:15 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
02:40:15 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
02:40:15 Join Kohlrabi [0] (
02:40:15 Join alexbobP [0] (
02:40:15 Join iq [0] (~iq@unaffiliated/iq)
02:40:15 Join @ChanServ [0] (ChanServ@services.)
02:40:15 Join Amec [0] (
02:40:15 Join amee2k [0] (
02:40:15 Join elcan [0] (user36@
02:40:15 Join CIA-7 [0] (~CIA@
02:40:15 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
02:40:15 Join jobec [0] (
02:40:15 Join scorche|sh [0] (
02:40:15 Join bug2000 [0] (~bug@unaffiliated/bug2000)
02:40:15 Join lostlogic [0] (~lostlogic@rockbox/developer/lostlogic)
02:40:15 Join user890104 [0] (
02:40:15 Join yosafbridge [0] (
02:40:15 Join Farthen [0] (
02:40:15 Join powell14ski [0] (
02:40:15 Join Utchy [0] (
02:40:15 Join ved [0] (
02:40:15 Join Rondom [0] (
02:40:15 Join Galois [0] (
02:40:15 Join TBCOOL [0] (
02:40:15 Join Slasheri [0] (miipekk@rockbox/developer/Slasheri)
02:40:15 Join Bawitdaba [0] (
02:40:15 Join maraz [0] (
02:40:15 Join jae [0] (
02:40:15 Join ThomasAH [0] (
02:40:15 Join Battousai [0] (~bryan@gentoo/developer/battousai)
02:40:15 Join Loto [0] (~nfs@xbmc/user/Loto)
02:40:15 Join pjm0616 [0] (~user@
02:40:15 Join zu_ [0] (
02:40:15 Join part [0] (part@
02:40:15 Join literal [0] (
02:40:15 Join bzed [0] (
02:40:15 Join burn [0] (
02:40:15 Join Stummi [0] (Stummi@rockbox/developer/Stummi)
02:40:15 Join Hadaka [0] (
02:40:15 Join BlakeJohnson86 [0] (
02:40:15 Join Gareth [0] (
02:40:15 Join preglow [0] (
02:40:15 Join ack [0] (
02:40:15 Join jordan` [0] (
02:40:15 Join Xerion [0] (
02:40:15 Join balintx [0] (
02:40:15 Join miceh [0] (
02:40:15 Join aevin [0] (eivindsy@unaffiliated/aevin)
02:40:15 Join Torne [0] (torne@rockbox/developer/Torne)
02:40:17 Quit factor (Ping timeout: 276 seconds)
02:41:56 Join timccc [0] (~timccc@
02:42:38 Quit Clipper (Quit: Page closed)
02:44:20 Join factor [0] (
02:45:53 Quit timccc (Ping timeout: 240 seconds)
02:52:33 Join timccc [0] (~timccc@
03:15:46 Join kadoban [0] (
03:37:03 Quit factor (Read error: Connection reset by peer)
03:39:38 Quit [Saint] (Disconnected by services)
03:39:38 Join factor [0] (
03:39:39 Join S_a_i_n_t [0] (S_a_i_n_t@
03:41:01 Quit moos (Ping timeout: 272 seconds)
03:50:30 Nick JesusFreak316_ is now known as JesusFreak316 (
03:51:11 Join moos [0] (moos@rockbox/staff/moos)
03:53:41 Quit GeekShadow (Quit: The cake is a lie !)
03:54:25 Quit JdGordon| (Quit: leaving)
03:55:25 Join fdinel [0] (
04:04:18 Join JdGord [0] (~jonno@
04:06:00 Quit fdinel (Read error: Connection reset by peer)
04:06:28 Quit [7] (Ping timeout: 260 seconds)
04:10:26 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:10:53 Quit JdGord (Quit: Bye)
04:14:27 Quit mortalscan (Ping timeout: 240 seconds)
04:14:53 Join CaptainKewl [0] (
04:19:50mooshehe, silly me :)
04:24:23 Quit mordocai (Ping timeout: 260 seconds)
04:26:21 Join JdGord [0] (~jonno@
04:28:27 Join Barahir_ [0] (
04:29:49 Quit Barahir (Read error: Operation timed out)
04:31:45CIA-7New commit by moos (r28886): Add "file size" to the track information screen of WPS. ...
04:33:48CIA-7r28886 build result: All green
04:37:57 Quit toffe82 (Remote host closed the connection)
04:40:14***Saving seen data "./dancer.seen"
04:40:45 Quit amiconn (Disconnected by services)
04:40:45 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:41:03 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:41:24 Quit designate72 (Ping timeout: 260 seconds)
04:41:53 Quit pixelma (Disconnected by services)
04:41:55 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:41:57 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:44:13CIA-7New commit by moos (r28887): Update translations reusing the translation of the previously deprecated string. ...
04:48:16CIA-7r28887 build result: 0 errors, 177 warnings (moos committed)
04:53:10CIA-7New commit by moos (r28888): Fix yellow, english.lang string was deprecated.
04:54:03 Join fischju [0] (
04:54:05 Quit fischju (Client Quit)
04:55:02CIA-7r28888 build result: All green
05:00:28 Quit madalu (Ping timeout: 276 seconds)
05:00:53 Quit froggyman (Quit: Ex-Chat)
05:01:54 Quit JdGord (Quit: Bye)
05:02:14 Join JdGordon| [0] (~jonno@
05:02:14 Quit JdGordon| (Changing host)
05:02:14 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
05:12:33 Quit factor (Ping timeout: 265 seconds)
05:13:12 Join factor [0] (
05:20:42 Join antil33t [0] (
05:22:17 Join JdGordon1 [0] (
05:25:49 Quit JdGordon| (Ping timeout: 276 seconds)
05:26:21CIA-7New commit by moos (r28889): Adjust a string translation in french, no more pure tag viewer since a while now. That's more a track information viewer.
05:28:05CIA-7r28889 build result: All green
05:30:38JdGordon1moos: 28887 looks wrong... shouldnt that "recording:" line just be removed?
05:31:28JdGordon1oh sorry, ignore me :D
05:31:38mooshehe :)
05:31:57moostoo slow! already in holidays? :p
05:40:09 Quit moos (Ping timeout: 240 seconds)
05:40:35 Quit JesusFreak316 (Ping timeout: 240 seconds)
05:44:47 Quit JdGordon1 (Ping timeout: 240 seconds)
05:45:00 Join moos [0] (moos@rockbox/staff/moos)
05:55:20 Quit n17ikh (Read error: Operation timed out)
05:55:51 Join n17ikh [0] (
06:25:15 Quit anewuser ()
06:40:16***Saving seen data "./dancer.seen"
06:59:02 Join froggyman [0] (~seth@unaffiliated/froggyman)
07:07:14 Join Judas_PhD [0] (
07:24:48 Quit kadoban (Remote host closed the connection)
07:27:16 Join kadoban [0] (
07:32:26 Join Horschti [0] (~Horschti@xbmc/user/horscht)
07:32:30 Quit CaptainKewl (Ping timeout: 264 seconds)
07:35:56 Quit Horscht (Ping timeout: 260 seconds)
07:44:45 Join mortalscan [0] (~mortalsca@
07:46:11 Quit kadoban (Remote host closed the connection)
07:47:14 Join kadoban [0] (
07:47:31 Quit Judas_PhD (Quit: This is a quitting message)
07:57:32 Quit kadoban (Remote host closed the connection)
07:59:01 Join kadoban [0] (
08:01:47 Join utanapischti [0] (~username@
08:33:17 Nick niekie_ is now known as niekie (~niek@CAcert/Assurer/niekie)
08:37:47 Quit mortalscan (Ping timeout: 240 seconds)
08:40:20***Saving seen data "./dancer.seen"
08:51:18 Quit utanapischti (Ping timeout: 272 seconds)
08:54:07 Join ender` [0] (
08:56:41 Join bertrik [0] (
08:56:41 Quit bertrik (Changing host)
08:56:41 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:57:08 Join JdGord [0] (~jonno@2002:7bf3:8c1f:1234:9221:55ff:fe70:3b02)
08:57:11 Quit JdGord (Client Quit)
08:58:28CIA-7New commit by jdgordon (r28890): Fix FS #11829 - %?xx<....> Crashes on targets where the %xx feature tag isnt avilable. rather hacky fix though better than crashing.
08:58:50 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
08:59:13JdGordon|^ that commit is a hack, but better than crashing....
08:59:35JdGordon|If i get time this weekend I'm going to rewrite the section so feature tags are handled differently to the conditionals so it isnt nasty anymore
08:59:51JdGordon|they should be handled as a no-op instead of a conditional check
09:00:26CIA-7r28890 build result: All green
09:12:18 Join utanapischti [0] (~username@
09:28:06 Quit JdGordon| (Quit: leaving)
09:30:30 Quit bertrik (Ping timeout: 240 seconds)
09:42:44 Join bmbl [0] (~bmbl@unaffiliated/bmbl)
09:45:49 Join thomasjfox [0] (
09:50:01 Join earcar [0] (
09:51:41 Join petur [0] (d408b802@rockbox/developer/petur)
09:55:45 Quit kadoban (Remote host closed the connection)
09:56:24 Quit earcar (Quit: bye)
09:56:58 Join kadoban [0] (
09:58:13 Quit timccc (Ping timeout: 255 seconds)
10:01:47 Join pamaury [0] (
10:01:47 Quit pamaury (Changing host)
10:01:47 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:06:07 Join timccc [0] (~timccc@
10:27:45 Nick shai_ is now known as shai (
10:34:30 Quit kadoban (Ping timeout: 240 seconds)
10:40:23***Saving seen data "./dancer.seen"
10:55:13 Join stoffel [0] (
10:55:58 Join kevku [0] (~kevku@2001:7d0:0:f000::135d)
11:03:30 Join dfkt [0] (
11:03:31 Quit dfkt (Changing host)
11:03:31 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:09:44 Join n1s [0] (
11:09:44 Quit n1s (Changing host)
11:09:44 Join n1s [0] (~n1s@rockbox/developer/n1s)
11:10:17n1shmm, the recent english.lang diffs will break the string order on some targets
11:10:42n1si wonder if we could come up with a better system for this stuff :/
11:11:12 Quit factor (Ping timeout: 265 seconds)
11:13:06LloreanWhat does breaking the string order break?
11:13:36n1svoice files from older revisions will play the wrong clip in some cases
11:14:04LloreanWouldn't the solution be to put a version number in voice files, and refuse to use them if they aren't compatible with the current .lang?
11:15:04LloreanIs there some reason we want people using old voice files?
11:15:45*TheSeven has evidence that the ipod classic OF uses OpenGL
11:16:23n1sthat is a solution but a) not very nice for blind users since voice will just stop working for them whithout any feedback ( the idea has come up to make for example the first clip an error like "Voice file out of date" that could play on version mismatches and b) would still require someone to manually update the version when breaking the order and it seems not many people realize when that happens
11:17:41Lloreann1s: Would it be possible to just compare the number of voice clips to lang strings and consider it broken if there's a mismatch? As for presenting feedback, maybe revert to "spell" when the voice file doesn't match so they can navigate the menus by spelled entries as a fallback.
11:17:57LloreanThat would of course require a subset of the voice file to be loaded, I assume
11:18:27n1sthat could work if the spelling clips are ok
11:18:36n1si don't remember if they are
11:19:00LloreanIs it possible to put them first and just say "don't mess with these"?
11:19:01 Join DerPapst [0] (
11:20:00n1sthe problem with the current system is that ID's get reused if we delete a clip so they are reserved by deprecating old ID's but if a target didn't use the id it never had it and what happened here is that all targets were marked as not having the clip (*: none)
11:20:12LloreanI know spelling is far from ideal, but it seems like falling back to spelled menu entries as soon as Rockbox detects even a *potential* mismatch is better than having it read random strings for entries.
11:21:11LloreanSo in the current situation, the total count of clips would've increased, no?
11:21:12n1syes, that would probably work for a fallback but the detection isn't easy
11:21:57n1stargets that didn't have the REC_SIZE clip are unaffected and the others have the order broken
11:22:53n1sjust comparing number of voice clips with number of lang ids will trigger when a voice file is just missing the most recent voice clip too
11:23:32LloreanWhile that won't mess with the order, that still leaves an entry in the menus unvoiced right? What happens currently when a user highlights an unvoiced entry?
11:24:23n1sLlorean: yes, currently i think it just isn't voiced, not sure though
11:24:27LloreanI mean, I think it'd be generally good to let people know they should update in that situation too, honestly.
11:24:29n1si imagine a scheme where id's are reserved globally and once deprecated, just not used
11:25:02n1syes, i agree
11:25:30LloreanI seem to recall rebuilding voice files if you hadn't cleared the cached clips was very quick anyway.
11:26:14LloreanI wonder how many people knowingly use the same voice file with multiple builds, and how many update when they download a new build even if they don't need to.
11:26:36LloreanIt'd be interesting to know how many this current situation is likely to even affect. :)
11:27:24linuxstbTheSeven: Is your work on the Classic documented anywhere? Have you published any working code?
11:27:44n1sthe "wrong clips are voiced" thing comes up from time to time but i have no idea how common it is, but anyway our current system isn't failing very gracefully
11:28:35TheSevenall the ipodnano3g code in here is currently shared with the ipod classic
11:28:47TheSeventhe ipodclassic folder only contains the classic-specific bits
11:30:41linuxstbDo you know if the three (?) generations of the classic are the same?
11:31:07 Quit kevku (Quit: KVIrc 4.0.2 Insomnia
11:32:13linuxstbSo the nano3g's LCD is the same as the Classic?
11:35:13TheSeventhere seem to have been small changes in some init sequences between the ipod classic 1g and 3g, and fourth lcd type was added at some point (there also are some 1g 160gb classics with the new lcd type, so i'd guess that this was done in the middle of 1g production)
11:35:31TheSeventhe nano3g and classic lcd driver are binary identical
11:36:00TheSeveni haven't tested any nano3g lcd code yet because we have no clue about its PMU
11:36:55linuxstbHow do you run your own code at the moment?
11:37:05TheSeveni have a DFU bootstrapper
11:37:39TheSeven(umsboot as a pwnage2.0 exploit payload)
11:38:04linuxstbSo the ipod needs to be attached to a PC to boot?
11:38:40TheSevenwe can also flash a bootloader
11:38:42amiconnn1s: The problem with better schemes is that they're significantly more expensive in terms of ram and binsize
11:39:12amiconnAs long as the rules are observed, wrong clips shouldn't happen with the current system
11:39:14TheSeveni have flashed iloader on mine, but if you just want to do non-invasive testing, you can bootstrap it all from RAM
11:39:46n1samiconn: but the rules arent' observed
11:40:15linuxstbTheSeven: OK, so you've ported iloader to the classic? What's embios?
11:40:38TheSevenembios is basically a small kernel + hardware abstraction layer used by my newer tools
11:40:44linuxstbSo I'm guessing you've done no actual work on a Rockbox port yet?
11:41:22 Join factor [0] (
11:41:38TheSeveni've plugged some things together (partly still nano2g code), and fixed it until it compiled, but it still doesn't link
11:42:18TheSeventhe iloader binary for the nano2g and classic is the same btw, the hardware differences are handled by embios
11:43:14TheSeveni have most of the important drivers up and running in embios, only audio is still missing
11:43:55 Join einhirn [0] (
11:44:16n1samiconn: including the unused ID's everywhere would cost only 3 bytes per string in the lang file and remove the whole feature change problem, which is the trickiest and least ovious one (i.e. a change in a target's config can change the string sincluded)
11:45:11TheSevenlinuxstb: the actual rockbox work i've done so far boils down to an afternoon of figuring out how to add a new port
11:45:19amiconnIt would also cost an additional 4 bytes of RAM per unused string on target
11:45:45amiconn(for the pointer)
11:45:55TheSevenbut the rockbox and embios driver interfaces are quite similar (embios, for example, borrows most of rockbox's storage and file system code)
11:46:07amiconnAnd this would mean the lowest end targets (with the least features) would suffer mosz
11:46:12linuxstbTheSeven: Yes, given you seem to have written most of the code, the rockbox port itself doesn't look like much more work. Do you think what you've done is committable? I may have time over Christmsa to work on it.
11:46:37TheSevennot quite...
11:46:43TheSeveni can send you a patch though
11:46:51amiconnBtw, this applies both to voice and .lng, just that the latter is much less of a problem because all .lng files are usually in sync since they're shipped as part of the default installation
11:47:07TheSeven"the rockbox port itself doesn't look like much more work" is what I thought, too, but which proved wrong
11:47:28TheSeven(maybe because i'm not really familiar with a lot of rockbox's internals)
11:47:32n1samiconn: yes, it's the voice files getting out of sync that's the problem
11:47:42linuxstbTheSeven: Yes, "not much work" if you know what to do...
11:48:34n1samiconn: and the current system isn't working, IMO because the changes that can break the string/clip order are not obvious unless you have dug around in this)
11:48:59linuxstbTheSeven: But sure, send me the patch, or maybe open a FS task?
11:54:43 Quit user890104 ()
11:54:56TheSeveni had a quick glance at the changes, and the only thing i seem to have copied straight from the nano2g is crt0.S
11:55:12TheSevenso this file will need a bunch of changes, i can do that later
11:55:39TheSevenif you give me something that compiles and links, i'll be happy to try figuring out why it doesn't boot :-P
11:56:09n1shmm, instead of storing a pointer to each string we could store an offsett into the loaded lang file which would only need to be 16 bits (although the lookup would be a little less efficient)
11:56:45n1sand would only work as long as lang files don't grow beyond 14k
11:57:17 Join user890104 [0] (
11:58:36 Quit einhirn (Ping timeout: 240 seconds)
11:59:31 Quit Guest16638 (Ping timeout: 264 seconds)
11:59:37*TheSeven found evidence of BDF fonts in the ipod classic OF
12:04:36 Join sasquatch [0] (~username@
12:06:14 Quit utanapischti (Ping timeout: 276 seconds)
12:06:27amiconnn1s: I've been thinking about verifying string ids on load, combined with only storing the valid ones as block start + block length
12:07:02amiconnIn the binary, obviously, as some sort of compression
12:07:23amiconnIf feature strings are grouped together, this should be fairly efficient
12:16:41n1syes, that would be better
12:24:11TheSevenhm, FreeType...
12:24:30TheSevenapple seems to use lots of open source code in their firmwares :)
12:30:42 Quit stoffel (Ping timeout: 264 seconds)
12:35:40TheSevenOpenSSL as well...
12:38:54 Quit dfkt (Read error: Connection reset by peer)
12:40:16 Join Stephencc__ [0] (
12:40:26***Saving seen data "./dancer.seen"
12:41:26Stephencc__the seven: you can see the classic uses open source as in teh credits it has to lsit them
12:42:43Stephencc__altho I tihnk you've wiped your apple OS correct ? maybe someone who has access to theirs can check it for you
12:55:32 Quit thomasjfox (Read error: Operation timed out)
12:56:01 Join Feisar [0] (
12:56:27 Nick Feisar is now known as Guest39819 (
12:56:33 Quit Stephencc__ (Quit: CGI:IRC)
13:00:27 Quit Guest39819 (Ping timeout: 240 seconds)
13:03:38 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
13:05:30 Quit factor (Ping timeout: 240 seconds)
13:09:03 Quit bmbl (Ping timeout: 240 seconds)
13:11:43 Join stoffel [0] (
13:12:46 Join bmbl [0] (
13:12:46 Quit bmbl (Changing host)
13:12:46 Join bmbl [0] (~bmbl@unaffiliated/bmbl)
13:22:54*TheSeven spots a postscript parser in the ipod classic OF...
13:28:17 Quit Topy (Read error: Connection reset by peer)
13:35:56 Quit evilnick_ (Ping timeout: 260 seconds)
13:41:13 Join evilnick [0] (
13:42:49 Quit bluebrother (Disconnected by services)
13:42:50 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
13:46:03pamaurya postscript parser ? To do what ?
13:46:32TheSevenno idea
13:46:40TheSevenprobably postscript font support or something
13:54:20 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:57:07 Join Feisar [0] (
13:57:34 Nick Feisar is now known as Guest36053 (
14:01:57 Quit Guest36053 (Ping timeout: 276 seconds)
14:02:30 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
14:14:08 Join factor [0] (
14:20:58 Join Topy44 [0] (
14:32:04 Quit petur (Quit: Page closed)
14:32:27 Join mortalscan [0] (~mortalsca@
14:35:02 Join thomasjfox [0] (
14:36:21 Join kevku [0] (~kevku@2001:7d0:0:f000::135d)
14:40:29***Saving seen data "./dancer.seen"
14:40:46 Quit factor (Read error: Connection reset by peer)
14:58:16 Join Feisar [0] (
14:58:38 Join factor [0] (
14:58:42 Nick Feisar is now known as Guest26518 (
14:59:48gevaertsthomasjfox: I tried your patch from the 11th yesterday. I still had audio dropouts when switching the LCD on and off (using the side sliding button). Did you manage to get rid of those with your later work?
14:59:53 Quit stoffel (Read error: Connection reset by peer)
15:00:24gevaertsOh, and welcome! :)
15:00:56 Quit mortalscan (Ping timeout: 260 seconds)
15:04:28thomasjfoxI'm just writing my port announcement to rockbox-dev including the latest patch
15:04:34thomasjfoxJust wait a few seconds ;)
15:09:02thomasjfoxgevaerts: The audio dropouts were solved by the gstreamer based audio backend
15:09:08gevaertsAh, excellent!
15:10:25thomasjfoxOne of the N900 support patches is already separated into FS #11834
15:10:40gevaertsYes, I'm looking at that one now
15:12:44kugeldo we really RGBSWAPPED in the sim if the target has it?
15:13:03thomasjfoxI tried it with the ipod nano 1g
15:13:08kugeli see
15:13:10thomasjfoxAnd it all looked messed up
15:13:31thomasjfoxI tried various players in the sim to make sure I didn't break the display code for those
15:13:51thomasjfoxkugel: I tried your new tagcache fix for the symlinks. Works like a charm!
15:13:56kugelperhaps it's faster to swap the framebuffer and use the RGB case
15:14:20kugelthomasjfox: it does, unless you compile the linux kernel on the same machine :P
15:14:21gevaertsDo we care about LCD speed for simulators?
15:15:05kugelwe surely didn't until now
15:15:18gevaertsI mean, optimising for the SDL app is needed of course, but the simulator is different I think
15:16:33gevaertsIf an optimisation happens to work for a sim too, fine, but I don't think it makes sense to spend time optimising them specifically
15:17:09thomasjfoxgevaerts: That's also the reason I gave up after two hours getting the RGB565SWAPPED case right
15:17:17gevaertsYes, exactly :)
15:17:28thomasjfoxI tried to tweak around the SDL_Pixelformat and the colors always ended up wrong...
15:17:55thomasjfoxAlso the SDL documentation states the "format" buffer is read-only.
15:18:03gevaertsthomasjfox: I think I'm going to add !defined(HAVE_LCD_SPLIT) to the optimised case. It's probably unlikely to ever happen on a colour display, but it will avoid surprises in the future
15:18:11thomasjfoxGood idea
15:18:48thomasjfoxkugel: You mentioned last time we could use native threads on the N900, too
15:19:20thomasjfoxkugel: There was something about a wait_for_interrupt() routine that is missing?
15:20:13kugelthomasjfox: I think you can also use kernel-android.c now (it's plain unix)
15:20:23thomasjfoxLet me have a look
15:20:43 Join CaptainKewl [0] (
15:21:15kugelbut maybe not, sdl has more sohpisticated interrupt simulation
15:21:29kugelandroid doesn't implement enable/disable_irq
15:25:00kugelah yes, some of my work is here: kugel-rb.git/shortlog/refs/heads/ucontext_thread">
15:28:36kugelit makes the sdl port use our native threads via swapcontext, but the wait_for_interrupt() related stuff is the same
15:29:14thomasjfoxwait_for_interrupt() is the same as before or the same as in the android port?
15:30:51thomasjfoxSomething different: I want to add a define for maemo.
15:31:02kugelif you use kernel-sdl.c then you need the one from my git repo, if you use kernel-android.c then take it from there
15:31:53thomasjfoxkugel: I'll try with kernel-android.c first
15:32:10 Join xavieran [0] (
15:32:40thomasjfoxThe N900 uses an ARM v8 is maemo5 based
15:32:56thomasjfoxI can imagine that rockbox would also run on the older N8xx series which is maemo 4 based
15:33:13thomasjfoxMaemo 4 won't have the phone related APIs like maemo 5 does
15:33:14kugelvery likely
15:33:21kugelthe android port also runs on 1.5
15:33:31krazykiti think soap was running a sim on his n8xx
15:34:03kugelthomasjfox: what do apps to shut up when a call comes in on maemo 4?
15:34:12thomasjfoxWould it be a good idea add a PLATFORM_MAEMO5 define?
15:34:14thomasjfox Later on someone could add a PLATFORM_MAEMO4 define or PLATFORM_MEEGO define
15:34:25kugelon android we solve api incompatiblities via reflection in the java layer
15:34:26soapkrazykit, not I
15:35:06kugelcan you dlopen() the phone related API? That'd allow runtime detection (and no seperate defines, and thus no seperate binaries)
15:35:53gevaertsI don't think separate binaries are really a problem for maemo
15:36:21gevaertsI guess distribution would use the repositories anyway
15:36:39 Quit Guest26518 (Ping timeout: 272 seconds)
15:37:58thomasjfoxgevaerts: Yes, exactly
15:38:15kugelthomasjfox: the vast majority will be same same for maemo 4 and 5 no? if yes then I'd probably prefer a single PLATFORM_MAEMO define together with an extra MAEMO_VERSION define
15:38:32thomasjfoxkugel: Sounds like a plan
15:38:59kugelotherwise you need CONFIG_PLATFORM & (PLATFORM_MAEMO4|PLATFORM_MAEMO5) everywhere
15:39:10thomasjfoxkugel: Yes, I wanted to avoid that if possible
15:39:22kugelunless of course you do #define PLATFORM_MAEMO (PLATFORM_MAEMO4|PLATFORM_MAEMO5) :)
15:39:29kugelthere's several ways to do it :)
15:39:29thomasjfoxas the port uses much of the SDL platform, it will get a bit messy, too
15:40:00 Join fml [0] (
15:40:28kugelprobably only in the beginning. in the mid and long term you probably want to replace sdl stuff with something tailored to unix/maemo
15:40:57gevaertsI'd guess display will always be SDL
15:41:52thomasjfoxOr we could use Qt later on. Though the SDL output is now reasonable fast
15:43:14gevaertsI'd say that unless we ever get to the point where we use third-party widgets, Qt isn't going to be faster than SDL
15:43:31gevaertsOther things might be easier of course
15:46:08kugelthomasjfox: is there a global volume on the n900?
15:46:58thomasjfoxbtw: fs#9431 looks like it can be closed after all the recent RaaA work
15:47:08kugelon android there is, and we don't try to override that with rockbox' own volume control: the volume buttons on each device control the global volume and the rockbox volume can only be configured from the interface. I quite like it that way
15:47:12thomasjfoxkugel: Yes, it gets modified by two hardware buttons
15:47:58thomasjfoxkugel: The gstreamer audio backend automatically modifies the global volume
15:48:44kugeldo you mean the volume bar in rockbox actually shows the global volume?
15:48:59thomasjfoxI have to check if there can be a separate volume for the audio stream without much hassle
15:49:14 Quit Strife89 (Quit: Leaving)
15:49:20kugelyou can use software volume control via dsp, android does that
15:49:50thomasjfoxthe built in nokia player also modifies the global volume
15:50:13 Join komputes [0] (~komputes@ubuntu/member/komputes)
15:50:56kugelpeople appreciate that on android because the hardware buttons don't allow fine grained control, while rockbox own control does
15:53:37thomasjfoxI have to check if a) I can easily change the volume for the audio stream only or b) how much CPU a software based volume change costs
15:53:38kugelwell, I'm slightly wrong. android doesn't use the software volume capabilities of rockbox (it's there, the ondas need it), but the android api has another layer of volume control (also essentially software based)
15:54:07kugelvolume change is the cheapest kind of dsp operation IIUC
15:55:55kugelisn't the n900 armv7?
15:56:42kugeldoes armv8 even exist?
15:59:12thomasjfoxIt says "ARM Cortex-A8-based" on the back. Have to check again
16:00:07thomasjfoxYou are right
16:00:09thomasjfoxIt's v7
16:00:27 Join LambdaCalculus37 [0] (
16:00:27 Quit LambdaCalculus37 (Changing host)
16:00:27 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
16:01:05linuxstbTheSeven: For some reason, is only giving me the first 65535 bytes of that patch when I download it (with firefox, curl or wget) - can you try another site?
16:02:19 Join ppawel [0] (
16:04:25kugelthomasjfox: I suggest you create a real maemo target in the target tree, and compile the sdl bits you need via firmware/SOURCES. That'd make your patch easier to review
16:05:02 Quit ppawel (Client Quit)
16:06:28thomasjfoxLike I said, it's still very messy
16:06:50thomasjfoxCan you help me a bit on the native arm threads?
16:07:21thomasjfoxI tried copying the kernel-android.c over the kernel-sdl.c and also the thread-android-arm.c
16:07:42kugelwhat issues do you have?
16:08:02fmlLambdaCalculus37: Hello. Could you please try and review FS #11827 ?
16:08:11thomasjfoxLooks like I have to revert that and adjust the firmware/thread.c "if (ANDROID)"?
16:08:15kugelyou woll need some stubs (or change change subs in system-target.h), the other sdl parts expect implementations android doesn't have
16:08:41kugelthread.c to include your thread-maemo-arm.c will also need adjustment, right
16:09:52thomasjfoxGuess it's best to create the maemo target tree first
16:10:08LambdaCalculus37fml: Added to my tasklist.
16:10:41 Join stoffel [0] (
16:11:00LambdaCalculus37fml: Does it work well for you so far?
16:11:50kugelgevaerts: did you give it a try already?
16:12:49kugelthomasjfox: I can have a look on your work on the DevConPubBerlin next week :) Maybe you want to join as well (if you're near Berlin)?
16:14:33gevaertskugel: sorry, no, A bit busy with other things now
16:15:06thomasjfoxkugel: Tuebingen is a bit far away ;)
16:17:04thomasjfoxgevaerts: If you try it: The application runs in fullscreen mode right now. By pressing CTRL+BACKSPACE you switch windows
16:17:20 Quit fml (Quit: CGI:IRC (Ping timeout))
16:17:23thomasjfoxI have to leave for a bit, a friend just came over
16:18:01gevaertsMight not be today. I'll try
16:19:00 Quit komputes (Remote host closed the connection)
16:22:05LambdaCalculus37thomasjfox: Maemo port? Nice. :)
16:22:52kugelthomasjfox: if you're using git you can publish repo on some hosting site so we can follow your progress more easily
16:40:30***Saving seen data "./dancer.seen"
16:51:55 Join madalu [0] (~user@unaffiliated/madalu)
16:57:25 Quit stoffel (Ping timeout: 240 seconds)
17:17:14 Join keanu [0] (~keanu@unaffiliated/keanu)
17:21:31 Join fdinel [0] (
17:23:51keanuI realize this isn't directly rockbox related, but I'm kinda hoping someone knows or can point me in the right direction - if I've got an sdhc card with a broken pin (pin 8, DAT1) is there any way to recover the data?
17:26:28 Quit LambdaCalculus37 (Quit: This computer has gone to sleep)
17:29:41 Quit benedikt93 (Quit: Bye ;))
17:30:55 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
17:31:56 Join mortalscan [0] (~mortalsca@
17:32:34 Join hrehf [0] (
17:33:28hrehfhrm, i can't find where i set the time/date on my sandisk clip+, yet there's a seemingly functional time display (just set to the wrong time, ofc) −− is this possible?
17:36:14hrehfaaaah, there's settings/system and system/ :o
17:36:15bertrikhrehf, some people argue that date / time is not a setting :P
17:36:54hrehfomg i never saw that and wass wondering why the clip+ wouldn't have it there (i googled it for generic and just saw system/ :D
17:36:59hrehfthanks ;-)
17:38:02gevaertsbertrik: it's the only logical conclusion!
17:38:24gevaertsSettings don't change unless you set them, therefore time can not be a setting!
17:39:09hrehfalso, rockdoom needs a better display on monochrome ;-)
17:39:48gevaertsOn monochrome, or on the clip? It doesn't look *too* bad on monochrome LCDs
17:41:14hrehfhm, on the clip+
17:41:45gevaertsThe problem is that the OLED display on those things is too fast, so temporal dithering doesn't quite work
17:42:14gevaertsOn LCD turning the pixels on and off at that speed makes them stick at some grey level in between
17:42:57hrehfthat's why in dark areas you only have random noise
17:43:19hrehfmaybe it should have an option to disable lighting
17:43:25*hrehf is interested
17:43:57gevaertsActually, I seem to remember someone on the forums who was working on a new .wad to help with this
17:44:00*gevaerts searches
17:44:28hrehfthis seems like it is something the engine should help with
17:44:32hrehfor does it?
17:45:02gevaertsLighting on its own isn't going to make much difference I'd expect
17:45:52hrehfwell my observation is that the bright hallway in the firsst doom2 lvl works well
17:46:05hrehfand alle the dark rooms don't
17:46:38hrehfyou could probably try to adjust contrast for each texture so that there's about 50% bright pixels in each one
17:47:45 Nick jepler- is now known as jepler (~jepler@emc/developer/pdpc.professional.jepler)
17:48:22hrehfsomeone should port quake ;-)
17:51:07 Join komputes [0] (~komputes@ubuntu/member/komputes)
17:52:32 Quit kevku (Quit: KVIrc 4.0.2 Insomnia
17:54:46 Quit kugel (Remote host closed the connection)
17:55:42 Join kugel [0] (
17:55:44 Quit kugel (Changing host)
17:55:44 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:06:37CIA-7New commit by jethead71 (r28891): mc13783 RTC: Handle years in a better way for the use of struct tm. Make the code less general because all years evenly divisible by 4 in the day ...
18:08:27CIA-7r28891 build result: All green
18:10:35 Join fml [0] (
18:11:51 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier.
18:12:04 Join WonTu [0] (
18:12:18 Part WonTu
18:13:38fmlLambdaCalculus37: yes and no. On the sim, it worked one time but not when I tried to repeat. I think it's because of dircache. But the current SVN code is wrong IMO since the suffix includes the trailing "/".
18:16:47 Quit fml (Quit: CGI:IRC)
18:26:13 Quit kugel (Ping timeout: 240 seconds)
18:29:06 Quit krazykit (Ping timeout: 276 seconds)
18:29:35 Quit komputes (Remote host closed the connection)
18:31:10 Join kugel [0] (
18:31:12 Quit kugel (Changing host)
18:31:12 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:38:52 Nick soap is now known as SoapGoneForChris (~soap@rockbox/staff/soap)
18:39:09 Nick SoapGoneForChris is now known as SoapGoneForX-Mas (~soap@rockbox/staff/soap)
18:40:34***Saving seen data "./dancer.seen"
18:45:33 Join krazykit [0] (~krazykit@
18:48:53 Join knopper [0] (
18:49:25knopperi have a small question how long does a rockbox port usually take?
18:52:59Tornedepends how much of the hardware is documented
18:53:08Torneand how similar it is to existing targets
18:53:13 Join kevku [0] (~kevku@2001:7d0:0:f000::135d)
18:53:40Torneif there's full docs for the chip and it has a similar dac/etc to something we already support, someone with experience could probably do it in a couple weeks to a functional level
18:53:53Torneif you have to reverse engineer everything and write a lot of new drivers, months or longer
18:54:09knopperi just at the wiki page of the sansa fuze, it it took nearly 2 years
18:54:35knopperso im hoping, the fuze+ port is available sooner ^^
18:56:33Tornethe fuze port didn't take two years, really, i don't think
18:56:46Tornethat may be how long froms omeone first looking at it until it was marked stable
18:56:54Tornebut the beginning of that time likely had little/no work actally being done
18:57:21knopperi think it is still disassembled or?
18:58:19Tornei don't know if anyone is even working on it at all yet
18:58:29Tornesomeone took one apart but that doesn't mean it's actually being ported
18:58:40Torneit's the same chip as another device that's being looked at, though
19:00:33knopperthe so called "Lyre project" stated that they have started working on it
19:01:34Torneno, they're working on a different board with the same chip
19:01:47Torneso the work is relevant, but it's not a fuze+ port
19:02:08Tornepamaury is still at a pretty early stage with looking at the bootcode
19:02:23knopperso lets say, rockbox works on this chip, than it would be easy to make it work on a fuze+ ?
19:02:55knopperso it is still many coding needed even if it works on it
19:02:57Tornei don't think we have a sensibel way to get anything installed on the fuze+ yet, we can decrypt its firmware updates but there's no unbricking process or anything
19:03:16pamauryespecially because I had to restart analysis from scratch. But it shouldn't be too hard to get the pins for most peripheral now.
19:03:20Tornealso, the other chips aren't gonna be the same; it helps a lot that the main system-on-chip is the same but there are some others
19:03:33Tornealso the lyre's port hasn't gotten very far yet either :)
19:03:40Torneso while effort can be shared, it's not already done
19:03:43pamauryHum, the manual says that if you press power for 20seconds, the firmware is restored, but it's unclear how it works
19:03:54Tornepamaury: yeah, i read that, seems odd
19:04:14pamaurysomeone tried, he said it installed an older version than it had, so perhaps there is a copy somewhere
19:04:23pamaury*he had
19:04:46knoppera port would be really nice because the fuze+ has the biggest display from all suported players until now
19:04:46pamauryyeah, of course
19:06:52pamaurydon't know, Torne might now such things ;)
19:07:04Tornewhat things?
19:07:10pamauryscreen size
19:07:13Torneoh. duno
19:07:34Tornei still intend to do an omap3 port sometime eee, and that'd run on pandora/n900
19:07:36Torne800x480 :)
19:07:43Tornewhere did those e's come from
19:07:44knopperah ok^^
19:08:00Tornein fact i have an omap3 board with HDMI out
19:08:04Torneso it could probably do 720p ;)
19:09:32thomasjfoxTorne: Check this
19:10:08 Join yelped [0] (
19:11:54yelpedDoes someone have a build enviroment set up, and is willing to compile rockbox for the Clip+ without radio?
19:11:56 Quit moos (Quit: Happy Christmas)
19:13:50pamauryyelped: why without radio ?
19:14:23Tornethomasjfox: no, not as an application
19:15:19thomasjfoxah ok
19:15:29kugelon the n900?
19:15:41Tornenot specifically on the n900, but it'll probably be little enough work that i might do it
19:15:56Tornethe pandora, mostly
19:16:03Tornethere's quite a few omap3 based devices out there
19:16:04kugelyou must be insane :-)
19:19:27yelpedBecause it's of no use to me and my friends. Just uses up space in the menu.
19:21:31pamaurywait a minute, I should be able to do that
19:21:47pamauryjust need to figure out how to disable the radio
19:22:45yelpedThanks. Just disable it in the Menu. I recall this being in IRC about 6 months ago.
19:23:16 Join kadoban [0] (
19:23:36pamauryI don't know much the iu code, if I can just tweak a define in config header, it will be faster
19:25:45kugelthomasjfox: did you see my message about publishing your git repo?
19:26:37thomasjfoxkugel: Yes, just have to find a nice place for it. Probably also the one
19:28:09 Quit yelped (Quit: CGI:IRC (Ping timeout))
19:28:09thomasjfoxToo bad rockbox doesn't have an official git server, then I could just put everything into a "maemo" branch
19:29:09pamauryarg, yelped just left
19:29:35 Quit knopper ()
19:34:40 Join earcar [0] (
19:37:31 Join Keripo [0] (
19:39:03 Quit earcar (Client Quit)
19:39:42 Join mischasworld [0] (
19:43:51 Join yelped [0] (
19:44:14 Quit kugel (Ping timeout: 240 seconds)
19:44:16yelpedSorry, my computer just crashed.
19:44:35pamauryyeah, I have your build, just need to find a place where to upload it
19:44:49yelpedIs Dropbox OK?
19:44:55 Quit thomasjfox (Quit: X-mas)
19:46:31pamauryWait a minute, I'm uploading on mediafire
19:46:47yelpedGreat. Thanks a lot!
19:47:01yelpedHow's the Fuze+ port coming along?
19:48:21pamauryI hope it's okay, I just remove the menu entry but didn't try it, so I encourage you to do so ;)
19:50:00yelpedActually I don't have a Clip+ with me now. I'll put it on my microSD card, and I'll copy it later. Thanks for everything.
19:51:02 Quit linuxstb (Ping timeout: 240 seconds)
19:57:07 Quit factor (Read error: Connection reset by peer)
19:58:22 Quit yelped (Quit: CGI:IRC (EOF))
19:59:59 Join factor [0] (
20:01:40 Quit factor (Client Quit)
20:02:01 Join factor [0] (
20:02:14 Quit mischasworld (Ping timeout: 240 seconds)
20:04:05 Join stoffel [0] (
20:08:08 Quit stoffel (Remote host closed the connection)
20:15:22 Quit hrehf (Quit: fasdf)
20:18:32 Join mischasworld [0] (
20:19:57 Quit fyrestorm (Quit: lamers envy me like they envy bill g -- main boot xp, just the way it should be!)
20:21:17 Join kugel [0] (
20:21:17 Quit kugel (Changing host)
20:21:17 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:31:19 Quit kugel (Quit: Yaaic - Yet another Android IRC client -
20:31:26 Join kugel [0] (
20:31:27 Quit kugel (Changing host)
20:31:27 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:31:50 Quit mischasworld (Ping timeout: 240 seconds)
20:37:08 Join linuxstb [0] (~linuxstb@
20:37:08 Quit linuxstb (Changing host)
20:37:08 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
20:40:40***Saving seen data "./dancer.seen"
20:42:23 Join mischasworld [0] (
20:43:21 Join mischa_ [0] (
20:43:52 Quit mischasworld (Remote host closed the connection)
20:45:02 Quit kugel (Ping timeout: 240 seconds)
20:48:02 Quit guymann (Ping timeout: 260 seconds)
20:48:56 Join guymann [0] (
20:49:01 Quit guymann (Client Quit)
21:01:50 Quit linuxstb (Ping timeout: 240 seconds)
21:05:17 Quit mischa_ (Quit: No Ping reply in 180 seconds.)
21:05:31 Join mischasworld [0] (
21:18:02 Quit mc2739 (Ping timeout: 272 seconds)
21:21:18 Join linuxstb [0] (~linuxstb@
21:21:18 Quit linuxstb (Changing host)
21:21:18 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
21:22:27 Quit CaptainKewl (Remote host closed the connection)
21:25:57 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
21:33:38 Join komputes [0] (~komputes@ubuntu/member/komputes)
21:34:36 Join LambdaCalculus37 [0] (
21:34:36 Quit LambdaCalculus37 (Changing host)
21:34:36 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
21:35:08 Quit mc2739 (Ping timeout: 272 seconds)
21:35:46 Quit LambdaCalculus37 (Client Quit)
21:36:39 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
21:38:04 Join Feisar [0] (
21:38:34 Nick Feisar is now known as Guest52307 (
21:40:09 Join linuxstb_ [0] (~linuxstb@
21:40:10 Quit linuxstb_ (Changing host)
21:40:10 Join linuxstb_ [0] (~linuxstb@rockbox/developer/linuxstb)
21:42:18 Quit n1s (Quit: Lmnar)
21:42:42 Quit linuxstb_ (Client Quit)
21:42:57linuxstbTheSeven: Thanks.
21:45:55 Quit bertrik (Quit: :tiuQ)
21:48:37 Join utanapischti [0] (~username@
21:49:42 Quit sasquatch (Ping timeout: 272 seconds)
22:02:50 Join guymann [0] (
22:08:30 Quit komputes (Remote host closed the connection)
22:10:38 Quit mischasworld (Ping timeout: 240 seconds)
22:16:51 Quit DerPapst (Quit: Leaving.)
22:17:09 Join Fuhrer [0] (75d3557b@gateway/web/freenode/ip.
22:35:44 Quit bmbl (Quit: Verlassend)
22:40:43***Saving seen data "./dancer.seen"
22:52:52 Quit kadoban (Ping timeout: 260 seconds)
22:52:54 Join T44 [0] (
22:56:14 Quit Topy44 (Ping timeout: 240 seconds)
22:56:26 Join mischasworld [0] (
22:59:15 Quit Fuhrer (Ping timeout: 265 seconds)
23:04:38 Quit mischasworld (Ping timeout: 240 seconds)
23:08:29 Join JesusFreak316 [0] (
23:15:51 Join mischasworld [0] (
23:21:50 Quit mischasworld (Ping timeout: 240 seconds)
23:22:14 Join anewuser [0] (anewuser@unaffiliated/anewuser)
23:29:09 Quit kevku (Ping timeout: 272 seconds)
23:32:03 Quit factor (Read error: Connection reset by peer)
23:32:39 Join factor [0] (
23:35:21 Quit factor (Client Quit)
23:35:46 Join factor [0] (
23:40:01 Quit pamaury (Remote host closed the connection)
23:51:20 Join mischasworld [0] (

Previous day | Next day