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 2009-10-02

00:00:02kugelconsider -fno-long-calls once the mapping is working fine :)
00:01:07TheSevenon the s5l8700, i can only map what's at address 0, and that needs to be iram, for the IRQs
00:01:32kugelisn't there an mmu?
00:01:49kugelwhat cpu is this?
00:05:59 Quit DerPapst ("Leaving.")
00:06:15markunkugel: no MMU: arm920t(or tdmi?)
00:06:54TheSevenbut wrapping on the address bus provides a nice fake for that
00:06:56markunah yes
00:07:44kugelI see, it has a PU and small caches
00:07:57TheSevennow let's check what's making the FAT driver panic as soon as i try to browse
00:08:31kugelthe graphic (ascii-art) on that page is confusing
00:09:04CIA-43New commit by mc2739 (r22870): Add myself to docs/COMMITTERS
00:09:52TheSevenkugel: they're talking about a special SoC containing both a 920t and a 940t on that page
00:10:17kugelyea, I suspected that
00:10:31kugelmc2739: welcome! ?0/
00:12:36TheSevenis there a quick way to define debug_printf to be splashf?
00:12:48TheSevenhow do i handle the varying argument count with such a thing?
00:13:23kugel#define debug_printf(...) splashf(HZ, ## var_args) ... or something like that
00:14:00kugeldebug-as3525.c has a similar define, took me some googling to figure it out
00:14:23 Quit mcuelenaere ()
00:14:27 Join stripwax_ [0] (
00:14:42TheSevenvarargs... and ##varargs
00:14:49 Quit bluebrother ("leaving")
00:15:17kugelI think the ## is just for taking the comma away (the one after HZ) if there's no additional args
00:16:59 Quit robin0800 (Remote closed the connection)
00:19:04kugelARMs model numbering is totally strange
00:19:39 Join JackWinter2 [0] (
00:20:24TheSevenah, it's "nvram.bin", that it's trying to write
00:20:51 Quit domonoky1 (Read error: 104 (Connection reset by peer))
00:21:21TheSevenheh, apple is really funny
00:21:37TheSevenif FTL init fails in their boot loader, they show a "low battery" screen
00:22:02kugelyour ftl is incompatible with the OF right?
00:22:27 Nick Unhelpful_ is now known as Unhelpful (n=quassel@rockbox/developer/Unhelpful)
00:22:43TheSevenwell, it's *supposed* to be compatible, but obviously trashed something again right now
00:23:17gevaertsTheSeven: go to an apple store and see how they respond "Every time the FTL init fails it shows the low battery screen! Can you fix that?"
00:24:04TheSeven"huh, something going wrong? must be the battery."
00:24:20gevaerts"no! I'm telling you! It's the FTL!"
00:24:31TheSeven"it's always the batteries"
00:25:10kugel"We'll gladly replace your battery for multiple dozens of bucks, but it may not solve your FTL problems"
00:25:27kugelalthough i doubt the costumer service has an idea of what an FTL is
00:25:43kugelpossibly thinking about some financial times london or so
00:26:22 Quit kugel (Remote closed the connection)
00:27:22TheSevenbut the devs obviously thought "ah well, no matter what's happening, blame it on the battery"
00:32:50 Quit stripwax (Read error: 110 (Connection timed out))
00:33:10 Quit JackWinter (Read error: 110 (Connection timed out))
00:35:54 Quit ender` (" The three most dangerous things in the world are a programmer with a soldering iron, a hardware type with a program patch a")
00:37:28 Quit Zagor ("Clint excited")
00:47:35 Join z35 [0] (
00:48:10 Quit hd (Client Quit)
00:49:53 Quit matsl (Read error: 110 (Connection timed out))
00:51:08stripwax_- what's "FTL"? Is that (one of) the reason(s) why we sometimes get low battery screens even when battery isn't low?
00:56:24 Join HellDragon [0] (
00:57:41 Quit Thundercloud (Remote closed the connection)
01:00:16 Join saratoga_ [0] (i=463f90ed@gateway/web/freenode/x-cccjdrtybueqyntf)
01:03:39 Join Strath_ [0] (
01:06:10***Saving seen data "./dancer.seen"
01:10:03 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
01:17:09 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
01:23:41 Quit chandoo ("Leaving")
01:28:49 Quit liar|netbook (Remote closed the connection)
01:33:41 Part toffe82
01:41:09 Quit stripwax_ ("")
01:42:25 Join KBH [0] (n=hbk@
01:49:21 Join Strife89 [0] (
01:59:25 Quit HBK (Read error: 110 (Connection timed out))
02:15:07 Quit saratoga_ ("Page closed")
02:27:30 Join Jarkinber [0] (
02:29:36 Quit Jarkinber (Client Quit)
02:35:11 Join CaptainKwel [0] (
02:37:35 Quit JdGordon (Read error: 104 (Connection reset by peer))
02:38:35 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
02:45:06 Join Strife1989 [0] (
02:45:46 Quit Strife89 (Nick collision from services.)
02:45:52 Quit xavieran (Remote closed the connection)
02:45:55 Nick Strife1989 is now known as Strife89 (
02:47:17CIA-43New commit by saratoga (r22871): Commit unrolled version of WMA's exponent decode taken from latest ffmpeg. Gives a 1% speedup @ 192kbps.
02:53:35 Quit Lss (Read error: 104 (Connection reset by peer))
02:54:18 Join AEnima1577 [0] (
03:06:13***Saving seen data "./dancer.seen"
03:14:15 Join FOAD_ [0] (
03:15:12AEnima1577Forgive me, I read in one of the notes somewhere back there, that this is not a discussion area, but I could not help myself: MUCH PRAISE, I just installed rockbox, and I have been very impressed and can't wait to take her around the block and see everything she is capible of, very excited and just wanted to say thank you for all of your efforts, everyone who works on this
03:20:15 Join Ebert_ [0] (
03:24:50 Quit panni_ ("( :: NoNameScript 3.81 :: )")
03:28:04Ebert_is it possible on an iRiver iHP to listen to a source through Optical-in ?
03:28:39Ebert_say, from my macbook into the iriver digitally, then out through headphone/line out
03:30:18 Quit FOAD (Read error: 110 (Connection timed out))
03:30:18 Nick FOAD_ is now known as FOAD (
03:52:14 Quit amiconn (Nick collision from services.)
03:52:18 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
03:52:20 Quit pixelma (Nick collision from services.)
03:52:22 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
03:52:25 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
03:52:42 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
04:04:55 Quit Strath_ ("Leaving")
04:05:54 Quit TheSeven (Nick collision from services.)
04:06:11 Join The_Seven [0] (
04:06:23 Nick The_Seven is now known as TheSeven (
04:08:44 Quit cg_ (Read error: 110 (Connection timed out))
04:23:48 Join likemindead [0] (n=XXXXXXXX@
04:26:08 Quit Rondom (Nick collision from services.)
04:26:18 Join Rondom [0] (
04:40:28saratogafinally got a clip build working that records all the playback state to disk
04:41:13saratogagoing to see if i can get some people to run it on the forums and give me the logs when it deadlocks
04:41:34JdGordonTheSeven: you're not allowed to say those sort of comments without more info!!!!....
04:41:56JdGordonI left my clip in seattle so cant run it.
04:42:12TheSevenJdGordon: Rockbox just created a nvram.bin and config.cfg :-)
04:42:33TheSevenand guess what the problem was coming from
04:43:08JdGordonis this with the OF still being able to read the disk?
04:43:30TheSevenjust browsed it in disk mode, and the files were fine
04:43:31JdGordonwell done!
04:43:58TheSeventhis means we need a controlled fashion to reboot rockbox now, as we need to properly unmount the FTL
04:44:40TheSevennorboot keeps acting picky if i simply reset it
04:45:01 Quit Llorean (Read error: 104 (Connection reset by peer))
04:46:42saratogaimpressive work
04:47:25TheSevenstill some bugs though. especially after unclean shutdown things start acting weird
04:49:05 Join Llorean [0] (
04:49:38TheSevennice, norboot seems to indeed commit leftover scattered pages after an unclean shutdown... but why does rockbox panic *sometimes* after one?
04:53:16JdGordonthere is shutdown handler in misc.c which you probably want to look at
04:55:36 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
04:56:51Dhraakellianjust installed r22864 on my Fuze last night (or this morning?)
04:57:00Dhraakellianscrolling is... a tad wonky
04:57:44Dhraakellianit occasionally backtracks one entry/line/whatever
05:01:36saratogaits related tot he screen issues too I believe
05:01:41saratogabertrik posted a patch on the tracker
05:02:43DhraakellianI was guessing that it had something to do with r22863
05:03:01Dhraakellianah, and someone posted mention of it in the Sansa AMS New Port thread
05:04:32 Quit likemindead (Read error: 104 (Connection reset by peer))
05:04:44 Join likemindead [0] (n=XXXXXXXX@
05:04:57 Quit evilnick_ ("TTFN")
05:06:03 Quit likemindead (Client Quit)
05:06:16***Saving seen data "./dancer.seen"
05:12:03 Quit amiconn (
05:12:03 Quit z35 (
05:12:03 Quit Paprica (
05:12:03 Quit fxb__ (
05:12:03 Quit Topy44 (
05:12:03 Quit rvvs89 (
05:12:03 Quit pcc1 (
05:12:03 Quit r00s (
05:12:03 Quit markun (
05:12:03 Quit ps-auxw (
05:12:03 Quit bzed (
05:12:03 Quit parafin (
05:12:03 Quit TheSeven (
05:12:03 Quit AEnima1577 (
05:12:03 Quit scorche (
05:12:03 Quit mc2739 (
05:12:03 Quit FlynDice (
05:12:03 Quit maraz_ (
05:12:03 Quit sbhsu (
05:12:03 Quit Bagder (
05:12:03 Quit GodEater (
05:12:03 Quit Llorean (
05:12:03 Quit _lifeless (
05:12:03 Quit saratoga (
05:12:03 Quit Rondom (
05:12:03 Quit CaptainKwel (
05:12:03 Quit crashd (
05:12:03 Quit daurn| (
05:12:04 Quit antil33t (
05:12:04 Quit tomers (
05:12:04 Quit niekie (
05:12:04 Quit gevaerts (
05:12:04 Quit AB3JU (
05:12:04 Quit Hadaka (
05:12:04 Quit rwong (
05:12:04 Quit Tuplanolla (
05:12:04 Quit uflops (
05:12:04 Quit Galois (
05:12:04 Quit meermanr (
05:12:04 Quit lyngaas (
05:12:04 Quit J-23 (
05:12:04 Quit DaveDavenport (
05:12:04 Quit HellDragon (
05:12:04 Quit AlexP (
05:12:04 Quit thegeek (
05:12:04 Quit DirtySkeMe (
05:12:04 Quit lostlogic (
05:12:04 Quit preglow (
05:12:04 Quit krazykit (
05:12:04 Quit Beta2K (
05:12:04 Quit sinthetek (
05:12:04 Quit avacore (
05:12:04 Quit DaCapn (
05:12:04 Quit rphillips (
05:12:04 Quit yosafbridge (
05:12:04 Quit CIA-43 (
05:12:04 Quit zu (
05:12:04 Quit rasher (
05:12:04 Quit goffa (
05:12:08 Quit polobricolo (
05:12:08 Quit ThomasAH (
05:12:08 Quit bubsy (
05:12:08 Quit KBH (
05:12:08 Quit Sajber^ (
05:12:08 Quit jon-kha (
05:12:08 Quit Slasheri (
05:12:08 Quit Torne (
05:12:08 Quit Shaid (
05:12:08 Quit w1ll14m|_ (
05:12:08 Quit aidy (
05:12:08 Quit blithe (
05:12:08 Quit tha (
05:12:08 Quit scorche|sh (
05:12:08 Quit dionoea (
05:12:08 Quit elcan (
05:12:08 Quit janne_oksanen (
05:12:08 Quit alexbobp (
05:12:08 Quit aevin (
05:12:08 Quit ChanServ (
05:12:08 Quit linuxguy3 (
05:12:08 Quit jfc (
05:12:08 Quit Rob2222 (
05:12:08 Quit kadoban (
05:12:08 Quit Dhraakellian (
05:12:08 Quit pjm0616 (
05:12:08 Quit jvd (
05:12:08 Quit jordan` (
05:12:08 Quit tarbo (
05:12:08 Quit JackWinter2 (
05:12:08 Quit advcomp2019_ (
05:12:08 Quit gtkspert_ (
05:12:08 Quit killan_ (
05:12:15 Quit BHSPitMonkey ("Ex-Chat")
05:13:42NJoinChanServ [0] (ChanServ@services.)
05:13:42 Join Llorean [0] (n=DarkkOne@rockbox/user/Llorean)
05:13:42NJoinRondom [0] (
05:13:42NJoinTheSeven [0] (
05:13:42NJoinamiconn [0] (i=quassel@rockbox/developer/amiconn)
05:13:42NJoinAEnima1577 [0] (
05:13:42NJoinCaptainKwel [0] (
05:13:42NJoinKBH [0] (n=hbk@
05:13:42 Join HellDragon [0] (i=jd@Wikipedia/HellDragon)
05:13:42NJoinz35 [0] (
05:13:42NJoinJackWinter2 [0] (
05:13:42NJoinAlexP [0] (n=alex@rockbox/staff/AlexP)
05:13:42NJoinPaprica [0] (
05:13:42NJointhegeek [0] (
05:13:42NJoincrashd [0] (
05:13:42NJoinscorche [50] (n=scorche@rockbox/administrator/scorche)
05:13:42NJoinbzed [0] (
05:13:42NJoinadvcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
05:13:42NJoinsaratoga [0] (i=9803c6dd@gateway/web/freenode/x-lklytxvqbpsofbwg)
05:13:42NJoin_lifeless [0] (n=lifeless@
05:13:42NJoinSajber^ [0] (n=Sajber@
05:13:42NJoindaurn| [0] (n=daurnima@freenode/staff/daurnimator)
05:13:42NJoinfxb__ [0] (
05:13:42 Join rasher [50] (n=rasher@rockbox/developer/rasher)
05:13:42NJoinpcc1 [0] (
05:13:42NJoinmarkun [50] (n=markun@rockbox/developer/markun)
05:13:42NJoinps-auxw [0] (
05:13:42NJoinr00s [0] (
05:13:42NJoinuflops [0] (
05:13:42NJoinrvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
05:13:42NJoinTopy44 [0] (
05:13:42NJoinRob2222 [0] (
05:13:42NJoinmc2739 [0] (
05:13:42NJoinkillan_ [0] (
05:13:42NJoinmaraz_ [0] (
05:13:42NJoinGodEater [0] (n=bibble@rockbox/staff/GodEater)
05:13:42NJoinsbhsu [0] (
05:13:42NJoinBagder [241] (n=daniel@rockbox/developer/bagder)
05:13:42NJoinFlynDice [0] (
05:13:42NJoinparafin [0] (
05:13:42NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
05:13:42NJoinpolobricolo [0] (
05:13:42NJointomers [0] (
05:13:42NJoinantil33t [0] (n=Mudkips@
05:13:42NJoinavacore [0] (
05:13:42NJoinDaveDavenport [0] (
05:13:42NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
05:13:42NJoinaidy [0] (n=aidy@
05:13:42Mode"#rockbox +o ChanServ " by
05:13:42NJoingtkspert_ [0] (
05:13:42NJoinbubsy [0] (n=bubsy@
05:13:42NJoinlinuxguy3 [0] (
05:13:42NJoinShaid [0] (i=adam@
05:13:42NJoinkrazykit [0] (
05:13:42NJoinTorne [0] (i=torne@
05:13:42NJoinjvd [0] (n=syscrash@poipu/developer/syscrash)
05:13:42NJoinBeta2K [0] (
05:13:42NJoingoffa [0] (n=goffa@
05:13:42NJoinniekie [0] (
05:13:42NJoinDirtySkeMe [0] (
05:13:42NJoinjfc [0] (
05:13:42NJoinalexbobp [0] (n=alex@
05:13:42NJoinjanne_oksanen [0] (n=janne@
05:13:42NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
05:13:42NJoinrphillips [0] (
05:13:42NJoinyosafbridge [0] (
05:13:42NJoinGalois [0] (
05:13:42NJoinThomasAH [0] (
05:13:42NJoinw1ll14m|_ [0] (n=nobody@
05:13:42NJoinpjm0616 [0] (n=user@
05:13:42NJointarbo [0] (n=me@unaffiliated/tarbo)
05:13:42NJoinjordan` [0] (i=gromit@
05:13:42NJoinkadoban [0] (
05:13:42NJoinDhraakellian [0] (
05:13:42NJoinHadaka [0] (
05:13:42NJoinAB3JU [0] (
05:13:42NJoinlyngaas [0] (
05:13:42NJoinDaCapn [0] (
05:13:42NJoinrwong [0] (
05:13:42NJoinmeermanr [0] (
05:13:42NJoinJ-23 [0] (
05:13:42NJoinaevin [0] (i=eivindsy@unaffiliated/aevin)
05:13:42NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
05:13:42NJoinelcan [0] (i=user36@
05:13:42NJoinsinthetek [0] (
05:13:42NJoinpreglow [0] (
05:13:42NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
05:13:42NJoinCIA-43 [0] (n=CIA@
05:13:42NJoinzu [0] (
05:13:42NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
05:13:42NJoinjon-kha [0] (i=jon-kha@
05:13:42NJoinblithe [0] (n=blithe@
05:13:42NJointha [0] (i=1038@
05:16:38TheSeventhat patch is based on current svn, and builds & works fine for me
05:16:48 Join likemindead [0] (n=XXXXXXXX@
05:17:00 Part likemindead ("FOSS FTW!")
05:18:22saratogawhats with the +/* PortalPlayer uses the USEC timer */ comment added?
05:18:41saratogaoh i see
05:21:21 Quit Strife89 ("If you hold a Linux shell to your ear, you can hear the C.")
05:23:08saratogaI posted a clip logging build in the test builds forum
05:23:41saratogai've been listening to the build for an hour now without a deadlock, but it does appear to be filling up the log as expected
05:24:19saratogasince different deadlocks seem to happen in different ways, maybe someone will manage to find something interesting
05:24:24saratogahmm less then an hour :)
05:25:15CtcpVersion from freenode-connect!freenode@freenode/bot/connect
05:25:46 Quit aidy (Read error: 131 (Connection reset by peer))
05:25:54 Join aidy [0] (
05:31:24TheSevenFS #10633
05:31:49TheSevenone can probably remove those MMU files again, as I don't use them any more
05:35:15*TheSeven is going to bed now, it's 0530am
05:44:14 Join xavieran [0] (n=xavieran@
05:45:32 Quit Horscht ("Verlassend")
05:53:42 Quit Ebert_ ()
06:09:03 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
06:19:07 Quit JdGordon ("Leaving.")
06:29:29 Join xavieran_ [0] (n=xavieran@
06:36:42 Quit xavieran (Read error: 110 (Connection timed out))
06:40:47 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
06:53:17 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
06:57:00 Quit AEnima1577 ("Leaving.")
06:59:02 Join cg_ [0] (
07:06:18***Saving seen data "./dancer.seen"
07:09:38pixelma_why does the "Spartan Black" theme show up on this page: ? It looks like it's made for a bigger screen, at least the preview does, the version you download has Ipod Photo in its name, so I assume it's a 220x176x16 version
07:21:11 Join Guest23293 [0] (
07:21:49 Quit n17ikh (Nick collision from services.)
07:21:51 Nick Guest23293 is now known as n17ikh (
07:22:05 Quit n17ikh (Client Quit)
07:22:33 Join n17ikh [0] (
07:24:32kkurbjunpixelma, I have a fix in hand that I think corrects the problems for brickmania for smaller screens that keeps the gameplay the same
07:24:43kkurbjunIt still need some bugs ironed out
07:25:00kkurbjunbut I ended up convertingthe whole plugin to fixed point and re-wrote all the collision code
07:25:36 Quit CaptainKwel (Remote closed the connection)
07:32:27 Quit togetic (Read error: 110 (Connection timed out))
07:32:56 Join Guest23293 [0] (
07:33:24 Quit n17ikh (Nick collision from services.)
07:33:26 Nick Guest23293 is now known as n17ikh (
07:40:32 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
07:42:02pixelma_kkurbjun: the fix in the tracker is already quite ok (except sometimes a "new ball" shoots off very quickly after it appeared) but all in all I find the game a bit too easy on my c200 now which is not due to the fix I guess but the changes underneath
07:42:08pixelma_have to go now though
07:42:55 Quit pixelma_ (" .")
07:56:37 Join matsl [0] (n=matsl@
08:01:20 Join stoffel [0] (
08:03:33 Quit HellDragon (Client Quit)
08:04:00 Join HellDragon [0] (
08:07:55 Join bertrik [0] (
08:14:11 Quit HellDragon (Client Quit)
08:14:26 Join flydutch [0] (
08:14:35 Join HellDragon [0] (
08:23:30 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
08:25:15 Join ender` [0] (
08:28:04 Quit gevaerts (Nick collision from services.)
08:28:15 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
08:35:48 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
08:35:51 Join T44 [0] (
08:39:53 Join MethoS- [0] (n=clemens@
08:43:13 Join Rob2223 [0] (
08:44:23 Quit BHSPitLappy (Remote closed the connection)
08:45:16 Quit bertrik ("De groeten")
08:49:55 Join esperegu [0] (n=quassel@
08:50:49 Quit aevin (Read error: 145 (Connection timed out))
08:51:06 Quit MethoS- (Remote closed the connection)
08:53:58 Quit Topy44 (Read error: 113 (No route to host))
08:54:22 Quit JackWinter2 (Remote closed the connection)
08:54:42 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
08:55:34 Join Thundercloud [0] (
08:59:32 Quit preglow (Read error: 110 (Connection timed out))
09:00:04 Join JackWinter [0] (
09:01:43 Quit Rob2222 (Read error: 110 (Connection timed out))
09:02:23 Join hd [0] (
09:06:21***Saving seen data "./dancer.seen"
09:06:36 Quit TheSeven (Nick collision from services.)
09:06:56 Join The_Seven [0] (
09:07:03linuxstbThe_Seven: Hi.
09:07:08 Nick The_Seven is now known as TheSeven (
09:09:14 Join reszerve [0] (
09:16:25 Quit HellDragon (Read error: 110 (Connection timed out))
09:31:33 Join petur [50] (n=petur@rockbox/developer/petur)
09:31:33 Join Trista887 [0] (
09:31:33 Quit feisar- (Remote closed the connection)
09:31:33 Join feisar-_ [0] (
09:31:33 Join Zambezi_ [0] (i=Zulu@
09:31:33 Quit ch4os (Remote closed the connection)
09:31:33 Quit tchan (Remote closed the connection)
09:31:33 Quit tmzt (Remote closed the connection)
09:31:33 Join ch4os [0] (n=ch4os@gentoo/user/ch4os)
09:31:33 Join tmzt [0] (
09:31:33 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
09:34:36 Quit tomers (Read error: 113 (No route to host))
09:38:22 Quit JdGordon ("Leaving.")
09:38:22 Quit notlistening (Read error: 104 (Connection reset by peer))
09:41:42 Quit _Tristan (Read error: 110 (Connection timed out))
09:42:43 Quit Zambezi (Read error: 110 (Connection timed out))
09:44:23 Quit Thundercloud (Remote closed the connection)
09:46:47 Join DerPapst [0] (
10:17:15 Join gigabit [0] (
10:17:33gigabitanyone else looking for beastpatcher like ,e
10:18:37gigabithey anyone know where i can find beastpatcher.exe for the gigabeast s
10:20:54 Quit linuxstb ("Leaving")
10:21:55 Join Bob_C [0] (
10:23:43gigabitanyone know where I can find beastpatcher.exe
10:27:06 Join vitamin_d [0] (n=ubuntu@
10:28:16vitamin_danyone here
10:28:55vitamin_dI need to find "Beastpatcher.exe"
10:29:34vitamin_dis there another way of installing rockbox on a Gigabeat S60 without beastpatcher
10:30:20 Quit gigabit ("CGI:IRC (EOF)")
10:32:57 Join nls_web [0] (i=www-data@
10:35:27vitamin_danyone know why they removed "beastpatcher.exe"
10:39:28 Join pamaury [0] (
10:40:20Lloreanvitamin_d: You can always compile it yourself.
10:42:04LloreanYou'll notice the Gigabeat S isn't one of the players we yet consider "well supported" and among other things this means precompiled installation tools may not always be available for such ports.
10:46:30vitamin_doh ok so i just follow the compile myself instruction thanks
11:00:22 Quit cg_ (Read error: 60 (Operation timed out))
11:01:18 Part vitamin_d
11:04:48 Join gapan [0] (
11:06:22***Saving seen data "./dancer.seen"
11:10:14Bob_CHow should I handle attributions when making significant changes to a module? Should I always leave the Copyright as original author?
11:14:21CIA-43New commit by markun (r22872): add gapless support for MP3 files encoded with iTunes
11:21:56 Quit nls_web ("CGI:IRC (Ping timeout)")
11:23:24DerPapstBob_C: i'd say leave the original author and add yourself
11:28:17Bob_Cok, thanks
11:29:25 Quit _lifeless (Read error: 110 (Connection timed out))
11:30:09 Quit reszerve ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
11:36:05 Join cg_ [0] (
11:46:12 Join MethoS- [0] (n=clemens@
11:49:47 Quit stoffel (Remote closed the connection)
12:01:09 Quit pamaury ("Parti")
12:13:16 Join funman [0] (n=fun@rockbox/developer/funman)
12:24:29Papricahey, someone can check the spider solitare plugin?
12:25:14 Quit MethoS- (Remote closed the connection)
12:25:28 Join MethoS- [0] (n=clemens@
12:26:07TheSevenIs FlySpray's jabber interface broken?
12:28:26gevaertsPaprica: one comment : many people don't want any more rock* or *box names
12:29:05CIA-43New commit by lenzone10 (r22873): Updated italian translation.
12:37:22Papricathe name is not the problem :)
12:37:41Papricado you have a new name?
12:38:31gevaertshow about just spider?
12:39:09Papricai think its have a problem with copyrights
12:44:29 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
12:46:24gevaertsfrom what I can find, it's a very old name, so I doubt that
12:46:46Tornecopyright doesn't cover names; you mean trademarks
12:47:03Torneand it's reasonable to assume that the name is generic at this point
12:50:23funmanTheSeven: probably
12:56:49 Quit esperegu (Read error: 104 (Connection reset by peer))
12:56:59 Join esperegu [0] (n=quassel@
12:58:09*TheSeven wonders who had the great idea of defining reg names like DATA_PTR, PARITY0, INT_CLR for the ECC peripheral...
12:58:41 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
12:58:43TheSeven*some* of them are prefixed with "ECC_", some aren't
12:59:41Bob_CIntel? They have C&D'd over the "i" prefix even
13:01:00TheSevenhi linuxstb :-) *waves*
13:04:44TheSevenlinuxstb: we seriously need to clean out that interrupt mess
13:04:48linuxstbHi - nice work with the FTL. I've only partly read the logs, what were the problems?
13:05:00TheSevenS5L8700 and S5L8701 don't seem to have much in common regarding ints
13:05:22TheSevenlinuxstb: there were dozens of problems.
13:05:41linuxstbSo maybe simply split system-s5l8700.c ?
13:05:52linuxstbi.e. create system-s5l8701.c
13:06:20TheSevenactually I'm wondering whay they are in there in the first place
13:06:25***Saving seen data "./dancer.seen"
13:06:52linuxstbWhy what are where?
13:06:53TheSevenI somehow feel like they would better be in firmware/export/s5l8700.h, in an ifdef
13:08:12TheSeventhe 2 biggest issues were memset() having a different argument order on rockbox vs. OFW (seems to have only affected writing though) and some weird behavior of that 512 vs. 2048 sector size emulation
13:08:45TheSeveni fixed that now by just running everything at 2048-byte sectors, and have only hit (and fixed) one bug that got triggered thereby
13:09:11TheSeventhe rest seems to be running pretty smoothly, so it's not as bad as we thought ("opening a can of worms")
13:12:03TheSevenBTW, I also read that CONTRIBUTING file, and I'm wondering to what extent the rules in there should be followed
13:12:47 Join evilnick_home [0] (
13:12:56TheSeveni.e. is it OK to have defines like FMCTRL1_DoReadData to enhance readability, or if we're really supposed to call them FMCTRL1_DOREADDATA for the sake of it
13:12:56 Join freqmod [0] (
13:13:36linuxstbTheSeven: Well, "we" make the rules, so they can always be changed if people agree. Personally I like your suggested convention, so am in favour.
13:14:50 Join robin0800 [0] (
13:16:31TheSevenand also I'm not too much in favor of 80-char lines, as this causes line breaks in too many things, which isn't good for readability either. Personally, I think 100 is a good tradeoff, and that's what I'm currently using in my code.
13:18:37TheSevenin there FTL, there are indentions up to 10 levels in some big init functions, which would only leave 40 chars for actual code. I know one could break out parts into subroutines, but that would disrupt reading it a lot, as it logically is one unit of code (which has quite some nested loops and ifs)
13:18:52TheSevens/there FTL/the FTL/
13:20:03*TheSeven will just ignore this "defines are all upper case" rule for now
13:20:56*TheSeven hopes nobody has used the "#define RSPaity0_0" (and lots of clones) yet
13:25:53TheSevenBTW what's the recommended way to remove implicit declarations? declare the functions immediately in front of them? all at the top of the file? an additional header?
13:26:42 Join _lifeless [0] (n=lifeless@
13:31:07AlexPBagder: Can you add the e200v2 manual to please?
13:31:49 Quit thegeek (Read error: 104 (Connection reset by peer))
13:31:53AlexPAnd the e200v2 manual to the daily manual builds of course :)
13:33:24 Join einhirn [0] (n=Miranda@
13:34:24markunTheSeven: I usually just move the function to the top
13:34:48markunwell, above the place where it is used at least
13:37:36TheSevengah, where does make get its dependencies from? there's a stray one to a nonexistant file right now...
13:43:51TheSevenare there any guidelines regarding the use of types like uint8_t?
13:45:43 Join Lynx_ [0] (
13:50:19 Nick fxb__ is now known as fxb (
13:50:32 Nick funman is now known as irma (n=fun@rockbox/developer/funman)
13:50:55TheSevenrefactoring that code will probably take all day...
13:51:14 Nick irma is now known as funman (n=fun@rockbox/developer/funman)
13:55:47 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
13:56:55pondlifeWhat's up with the build table? Or is that now normal mid-build or somesuch?
13:57:36gevaertsit only shows not-fully-green columns these days
13:57:52 Join panni_ [0] (i=hannes@
13:58:26pondlifeThe Recorder - Sim column is fully green...
13:59:09gevaertsZagor: bug!
14:00:29pondlifeZagor: Did you see my ancient proposal for an all-in-one build table - maybe still available at">
14:02:18ZagorI don't see how that is simpler. it's still a screenful of data.
14:02:23pondlifePuts n builds across the screen - so horizontal scrolling is only needed if you want to look back in time. Combines the deltas and build results into one.
14:02:44pondlifeIt's not simpler, particularly...
14:02:59pondlifeJust more compact
14:02:59 Quit antil33t (Read error: 104 (Connection reset by peer))
14:03:06 Join antil33t [0] (n=Mudkips@
14:04:33pondlifeEach entry is target/bootloader/sim-delta
14:10:06 Quit linuxstb ("Leaving")
14:15:36*DerPapst likes pondlifes table
14:24:44 Join cpuguest [0] (i=dontryth@
14:27:40 Join thegeek [0] (n=nnscript@
14:31:54 Quit xavieran_ (Read error: 104 (Connection reset by peer))
14:32:46 Join xavieran [0] (n=xavieran@
14:35:17 Nick Zambezi_ is now known as Zambezi (i=Zulu@
14:46:49TheSevenmarkun: I meant "warning: function declaration isnât a prototype", not implicit declarations
14:50:05TorneTheSeven: are you missing the void for functions iwth no args?
14:50:12Torneint foo(); is not a prototype
14:50:19Torneint foo(void); is
14:50:30 Quit robin0800 (Remote closed the connection)
14:50:42Zagorpondlife: sorry for being grumpy before. I agree yours is a better table.
14:53:15 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
14:54:30 Join Talu_Foxtrot [0] (
14:54:35Talu_Foxtrothi everyone
14:58:50 Join mt [0] (n=MTee@rockbox/developer/mt)
15:00:55Talu_Foxtrotis anyone able to help me with an installation problem for Rockbox?
15:01:45 Quit MethoS- (Remote closed the connection)
15:03:28linuxstbTalu_Foxtrot: Just explain your problem.
15:03:32TorneTalu_Foxtrot: not unless you explain wha tit is
15:04:28Talu_Foxtroti have the latest exe and an ipod classic 80gb, but it wont allow me to continue as it sais "cannot detect device"
15:04:56Talu_Foxtroti have the selected device filled in
15:06:26***Saving seen data "./dancer.seen"
15:06:39linuxstbRockbox doesn't work on the Classic.
15:06:49Talu_Foxtrotah oks
15:07:09Talu_Foxtrotany recomendations for similar firmware that does per chance?
15:07:22TorneThere is no similar firmware to rockbox
15:07:31TheSevenTalu_Foxtrot: there is nothing besides the apple firmware for these
15:08:01Talu_Foxtrotyeesh, i am doomed >.>
15:08:14Talu_Foxtrotwell thanks for the help
15:08:33Talu_Foxtrottake care all *wanders off into the netterwebs*
15:08:35 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
15:08:50 Quit Talu_Foxtrot ("Leaving")
15:09:51pondlifeZagor: No grumpiness was received or implied ;)
15:10:12pondlifeIt was just an idea, probably a bit of a nightmare to code up.
15:10:37ZagorI like that it does away with the need for images in the top header
15:10:44pondlifeAlthough I would say I prefer to see the green - a bit of positive feedback
15:12:11pondlifeAnd that orientation was more scrollwheel-friendly.
15:14:34 Quit esperegu (Read error: 113 (No route to host))
15:16:30 Nick uflops is now known as mikroflops (
15:20:09 Join esperegu [0] (
15:21:33 Join cool_walking_ [0] (
15:32:04amiconnTheSeven: FMCTRL1_DO_READ_DATA
15:32:49amiconnRegarding prototypes - I usually put all required ones at the top, but both styles are used in rb
15:34:43 Quit funman ("week-end!")
15:35:10TheSevenamiconn: 80 chars per line is really tough with that kind of naming convention
15:37:44*amiconn almost always adheres to the 80 column limit without problems
15:38:41TheSevenresults in useless crap like that for me:
15:38:43TheSeven uint32_t max = (*ftl_nand_type).pagesPerBlock;
15:38:44TheSeven for (k = 8; k < max; k += 8)
15:41:30TheSevenheh, is use of "goto" ok?
15:45:09kkurbjunpixelma: the problem with the patch in the tracker is that it is not a /proper/ fix, a couple of the ball x speeds will calculate out to 1 which makes the game off - this could also potentially make bricks unreachable - I don't think that is currently the case, but if you don't have enough spread across the different x/y speeds depending on the screen size there may be areas that are untouchable.
15:45:12kkurbjun The other problem in Brickmania as it is in svn is that if the ball speep is too high the collision detection starts to fail. I converted most of the collision checks to see if there is an intersection between two lines. This will allow the ball speed to be drastically increased without the collision detection failing. Ideally this paves the way for a higher difficulty setting as I find the game too easy on any screen size.
15:46:39pixelmait got way easier with your unification, especially on the small screens I tested with (c200 and Ondio)
15:46:49kkurbjunI also noticed that the game for some reason does not properly scale for very small screens which is something that I am looking at
15:46:52TheSevenlinuxstb: what further steps are required to make plugins and codecs build and work?
15:48:07linuxstbTheSeven: Not much at all. Tremor fails to compile (due to a need for long calls, which I haven't investigated), and libfaad also failed due to a #define clash with your NAND driver (but you may have changed that by now).
15:48:27linuxstbAnd I think there were also some name clashes with one or two other plugins. Let me upload my patch to "apps/"
15:48:54TheSevenwhich define has been clashing?
15:49:19linuxstbI forget now - I think "ECC" was one, but maybe one or two other 3-letter defines.
15:49:45 Quit gapan ("Leaving.")
15:50:04*TheSeven thinks he hasn't used such short names in global scope and the long ones should all be prefixed with ECC_, NAND_ or FTL_
15:50:25kkurbjunpixelma: I think the issue on those screens is the height to width ratio is different from the H300. I am going to try scaling the x/y speeds with different ratios to see how the difficulty works out
15:50:38pixelmakkurbjun: what do you exactly mean with "scale"
15:51:23linuxstbTheSeven:"> is my patch. Seems I was wrong about libfaad (AAC), but maybe I fixed that locally.
15:52:35linuxstbSo I think the "png" plugin was the only one which clashed.
15:52:51linuxstbA couple of other plugins made false assumptions - hence the addition of some more #ifdefs
15:52:53TheSevenpng and vorbis...
15:53:14kkurbjuneverything is currently based on the H300 as a reference speed. All of the players (should) run at a fixed amount of time per screen rendering. As the screen size varies the pixel width of the paddle, and bricks should vary. Consequently, if you are moving the ball the actaul number of pixels you move it should also vary based on the screen size if you assume that each frame has a fixed rendering time
15:53:30linuxstbTheSeven: vorbis failed due to long calls. You'll see the error if you try to compile it.
15:53:41TheSeveni'll look into this once i have reworked my ftl code, which will probably need some more hours
15:53:49pixelmakkurbjun: ok, my one wish is though that it still uses as much screen space as possible, not wasting columns or so just to maintain the ratio. It is small enough as it is - and I don't mind little differences between targets, the random users has just one target anyway
15:54:10 Join kugel [0] (n=kugel@rockbox/developer/kugel)
15:54:42linuxstbTheSeven: I can't remember what more is needed for audio - probably just setting the clocks correctly. Rockbox's playback engine outputs at 44.1KHz, so that's the only samplerate we need to support to get audio playback. Some plugins may use other samplerates (if available) though.
15:54:52 Join evilnick [0] (i=0c140464@gateway/web/freenode/x-vssmtmmrwwvlkrzw)
15:55:25 Join AndyI [0] (n=pasha_in@
15:55:29TheSevenwas the ofw running at 48kHz?
15:55:52kkurbjunbefore, the game used the same step size across all of the screens, this caused problems on really large screens because the game was way too slow and on small screns it was unusually hard because the game was too fast. The hack was to increase or decrease the rendering time per frame, but that only worked to a point, it's not reasonable to expect a target with a very large screen to render a frame in a very small amount of time
15:56:15linuxstbTheSeven: I can't remember... It may have been.
15:56:15kkurbjunpixelma: the ball and brick positions should be the same
15:56:27kkurbjunlikewise for the text and other renderings
15:57:40kkurbjunpixelma: having unique defines for every target is a nightmare for maintenance when it can be done the same across all the targets
15:58:01pixelmaI value "readabilty" much higher
15:58:09kkurbjunI think that is one of the worst things about doing a new port is the plugins
15:58:13kkurbjunreadability of what?
15:58:37 Part cpuguest
15:58:57kkurbjunif you mean the code, I think brickmania was the worst offender, it was barely maintainable, no comments, long strings of comparison with poorly named variables, etc...
15:59:48linuxstbTheSeven: Does your NAND code now deal directly with 512-byte sectors? I see you've removed my hack from nand_read_sectors()
15:59:49pixelmae.g.the small H10 (with the square 128x128 screen) has the same brick width but one pixel higher bricks than the c200 (132x80 screen) so that you use the screen better
16:00:15kkurbjunthe brick sizes are not changing
16:00:42TheSevenlinuxstb: no, i propagated that up, and tuned the FAT driver to directly work with 2048 byte sectors which is better from almost any aspect
16:01:16TheSeventhis only downside is that you can't mount media with a lower sector size with that build, which obviously won't hurt
16:01:27pixelmakkurbjun: just what I imagined from your explanations
16:01:32linuxstbTheSeven: Did you forget to include that in your patch? Or am I blind?
16:01:47TheSevenanother thing that needs to be investigated is how usb_storage reacts on this, which is to be done later
16:01:53kkurbjunok, you compare what I will commit to what is in there now and tell me it is worse
16:02:03TheSevensee fat.h and config-s5l8700.h
16:02:14TheSevenconfig-ipodnano2g.h of course
16:02:17linuxstbfat.h isn't in your patch. Was it changed?
16:02:53kkurbjun(that was a ridiculous comment by the way - without any reference at all)
16:03:02linuxstbTheSeven: Similarly, config-s5l8700.h isn't there...
16:03:22TheSevenheh, funny. svn too dumb to properly export that patch?
16:03:24TheSevenneed to look into this
16:03:41 Join froggyman [0] (
16:04:32TheSeveni basically did a #define SECTOR_SIZE 2048 in config-ipodnano2g.h and the following in fat.h:
16:04:32TheSeven+#ifndef SECTOR_SIZE
16:04:32TheSeven #define SECTOR_SIZE 512
16:04:32DBUGEnqueued KICK TheSeven
16:07:28 Quit AndyIL (Read error: 110 (Connection timed out))
16:07:46linuxstbTheSeven: Your diff should be done at the root of the rockbox source - e.g. "svn diff firmware > mypatch.diff". It's looks like you did it inside firmware/target/
16:08:33TheSevenhmm, would explain the problem, but I can't imagine how that happened.
16:08:57 Join liar|netbook [0] (n=liar@
16:09:02linuxstbI assume you also changed the sector buffer used to read the MBR? IIRC, that was hard-coded to 512 bytes.
16:09:19TheSevenyep, I changed that to SECTOR_SIZE
16:10:11 Part LinusN
16:16:34 Join preglow [0] (i=thomj@rockbox/developer/preglow)
16:17:41linuxstbTheSeven: OK. Well, it sounds sensible to me, but I don't know the internals of FAT (or Rockbox's driver). Does you patch change the code for other targets?
16:18:11TheSevenit shouldn't, as it's only preprocessor-level stuff
16:18:25TheSeveni just added the possibility to override the sector size in config-ipodnano2g.h
16:19:05TheSeventhe code seems to be already working fine for the big sectors
16:23:20 Quit froggyman ("CGI:IRC")
16:25:25 Quit matsl (Read error: 145 (Connection timed out))
16:28:37 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
16:28:48 Quit DerPapst (Read error: 104 (Connection reset by peer))
16:29:27 Join DerPapst [0] (
16:31:41 Quit linuxstb ("Leaving")
16:31:59 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
16:32:33 Quit cool_walking_ (Read error: 104 (Connection reset by peer))
16:33:35 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
16:34:01 Join cool_walking_ [0] (
16:34:31 Quit GeekShadow (Read error: 54 (Connection reset by peer))
16:45:28 Join chandoo [0] (
16:45:36 Join Grahack [0] (
16:46:53 Join toffe82 [0] (n=chatzill@
16:47:39TheSevenOK, my files are building without warnings now
16:47:57TheSeven(and they should at least adhere to most of the code guidelines now)
16:50:33linuxstbHave you tried plugins yet?
16:51:18 Quit Grahack ("Leaving.")
16:58:13TheSevennot yet
16:58:28TheSeveni'll now first split up that patch before touching the tree again
16:59:17 Quit Sajber^ (Read error: 131 (Connection reset by peer))
17:01:21 Quit Zagor ("Don't panic")
17:06:27***Saving seen data "./dancer.seen"
17:09:53 Join faemir [0] (n=faemir@
17:16:59 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
17:17:52linuxstbTheSeven: Is your patch safe to commit? I don't really have time to test it, but if it looks good, and doesn't break any other target, I'll probably just commit it blindly.
17:18:07linuxstb(e.g. does it change the Meizus?)
17:18:47CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
17:18:47*linuxstb remembers that that patch includes his own unified ipod clickwheel driver, which may need some more cleanup/abstraction (and testing on an old ipod)
17:20:18TheSevenlinuxstb: just uploaded new patches at FS #10633
17:21:00TheSeveni have modularized it now, if you aren't sure about the clickwheel driver, you can easily strip that out in patch #4
17:22:18TheSeventhere are in fact some code changes in crt0.S that will affect the meizus (protection unit setup), but this should effectively do the same thing as the old code
17:22:47TheSeven#1, #2 and #5 should be absolutely safe
17:23:28saratogathe FTL on the Meizus is completely different then the nano2g right?
17:23:52gevaertsI still feel that #1 should be done in a more general/integrated way, but I don't really know how...
17:24:02gevaertsso ignore me on that
17:24:19TheSevengevaerts: we can still do that at a later point by just moving that define
17:24:34TheSevensaratoga: I haven't seen any meizu data yet, but from what bertrik told me, it's pretty much different
17:25:49TheSevengevaerts: I just need to get #1 in now to make the FTL work, and this ist the least invasive way. I wouldn't dare to dig deeper into storage right now.
17:27:49saratogaseems reasonable, the other Meizu and Samsung ports don't actually work anyway
17:28:27saratogaTheSeven: if you get around to registering a wiki account, updating this table might be a good idea too so that other developers are aware of your progress:
17:28:52TheSevensaratoga: I registered one yesterday, but don't think I have write access yet
17:29:14saratogawhats your account
17:31:03TheSevenlinuxstb: what do you think about the patches?
17:31:40saratogaTheSeven: should be done
17:31:45saratogawiki access that is
17:32:09 Quit daurn| (Read error: 101 (Network is unreachable))
17:34:49 Join Horscht [0] (
17:35:51linuxstbTheSeven: I'm not sure about your patch #2 - I would like the opinions of others regarding your choice of mixed-case for #defines (amiconn seems to object, if I understand his messages earlier in the logs)
17:36:25TheSevenwell, this should be easy to fix...
17:37:16linuxstbIs "RSPaity0_0" etc how the datasheet refers to them? Those are register names, not bit defines.
17:37:35linuxstbOops, they were already there ;)
17:37:44linuxstb(and you fixed the typo)
17:38:13polobricoloTheSeven: i get a lot of SECTOR_SIZE redfinition warnings. is it normal ?
17:38:17linuxstbPersonally, I'm not objecting - I just don't want to upset others if anyone feels strongly.
17:38:42linuxstbTheSeven: I guess we can just commit, and change if anyone shouts...
17:39:28*linuxstb needs to checkout a clean SVN
17:40:21*kugel would object
17:40:37TheSeveni would suggest committing them in the order they are numbered - seems to make most sense to me
17:40:55TheSevenkugel: regarding mixed-case defines?
17:43:02TheSevenkugel: so how would you handle things like that? #define FMCTRL1_ClearWFIFO (1 << 6)
17:43:23kugelall uppercase of course
17:43:43linuxstbkugel: You don't think it's an advantage to separate the name of the register with the name of the bitfield?
17:43:48 Join Strife89 [0] (
17:44:16linuxstb(as a general convention - e.g. if the register name had a _ in, it would be ambiguous)
17:45:05kugellinuxstb: what case are you talking about?
17:46:42linuxstbWhat do you mean?
17:46:52kugel"kugel: so how would you handle things like that? #define FMCTRL1_ClearWFIFO (1 << 6)"
17:47:02kugelerr, "kugel: You don't think it's an advantage to separate the name of the register with the name of the bitfield?"
17:48:12kugelit's not only mixed case, but also camelCase which is against our guidelines too
17:48:31linuxstbI'm saying that if used "FMCTRL1_CLEAR_W_FIFO" for that name, then that would be ambiguous for cases where the register name had an underscore.
17:49:14TheSevenlinuxstb: honestly, i would rather go for FMCTRL1_CLEARWFIFO
17:49:59kugellinuxstb: I only see FMCTRL0
17:50:04kugelas a register define
17:50:21*linuxstb just remembers having similar problems with other #defines (not in this file), and TheSeven's convention would have been useful then.
17:50:31linuxstbI'm not talking specifically about TheSeven's patch.
17:51:25kugelbut I am, and I can't see the problem you're asking me about
17:51:40kugeldefines with underscore == registers, without == bits? I never head of that
17:51:53linuxstbkugel: No, that's not what I'm saying.
17:52:31 Join robin0800 [0] (
17:52:51*kugel wouldn't even do bit defines for bits that are only used in for a *single* register
17:52:56linuxstbI'm saying that if the name of a register had an underscore (e.g. looking in s5l8700.h there is an ADM_CONFIG register), then the name could be less meaningful if it was something like ADM_CONFIG_BIT_MASK_VALUE
17:53:08kugelat least not in such a global header file
17:53:26*linuxstb can't be bothered to argue, so if TheSeven is happy to change...
17:54:31*TheSeven just spotted some files that were still missing in one of the patches
17:55:49 Quit petur ("burps")
17:56:16TheSevenkugel: i put them there because they are *different* for S5L8700/S5L8701, but the rest of the code can probably be shared
17:56:26 Quit _lifeless (Remote closed the connection)
17:56:43 Join _lifeless [0] (n=lifeless@
17:58:17kugelour guidelines are there for a reason, if ambiguity is a problem another solution should be found (such as adding _REG to all registers
17:58:50AlexPThe guidelines can surely change if something comes up that wasn't thought of at the time
17:58:57AlexPOr wasn't a problem at the time
17:58:57TheSevennot that they would have been followed in other places...
17:59:09 Join funman [0] (n=fun@rockbox/developer/funman)
17:59:10kugelor, as I would do, don't define bits that need the corresponding reg in its name because it's only used for that
17:59:25kugelTheSeven: that's no excuse in my opinion
17:59:47TheSevenkugel: so you would hardwire these things?
17:59:53AlexPNote that I'm not commenting on this at all in particular, but the guidlines say so isn't much of an argument without saying why they say so
18:00:03*gevaerts agrees with AlexP
18:01:35gevaerts"our guidelines are there for a reason" should only be said if you're prepared to state that reason :)
18:01:44AlexPand defend it :)
18:02:30kugelall uppercase defines is clearly not just our guideline. it's a very widely followed convention everywhere
18:02:49*linuxstb welcomes TheSeven to the harmonious Rockbox community ;)
18:03:34TheSevenare rockbox headers somehow protected against double inclusion?
18:03:49kugelwell, usually
18:03:51linuxstbTheSeven: You should do that yourself in each .h
18:04:00LloreanDon't our guidelines also say that exceptions are commonly made for code from other sources to maintain existing terms/names/etc?
18:04:24TheSevenlinuxstb: you mean such IFNDEF...DEFINE constructs?
18:05:27 Quit esperegu (Read error: 113 (No route to host))
18:06:11TheSevenLlorean: Yep, actually I could just refuse to port it at all :-P
18:06:12funmanmy attempts at adding greylib support to yh920 causes a crash in mpegplayer' MC_put_o_16_align0 ; and I remember seeing crashes in this function several times in the past. Does this remind you something ?
18:06:16kugelanyway, it clearly falls under the "When changing code, follow the code style of the file you are editing." rule also (i.e. I don't want to block the commit)
18:06:34kugelbut it should be changed to all uppercase in some separate commit IMO
18:06:44funmanI was supposing cache corruption, the PP502x don't use cp15 to manage i/d caches but a custom controller
18:06:57linuxstbTheSeven: Yes. that's how it's normally done. Do you know a better/different way?
18:07:07 Join Lss [0] (
18:07:21TheSevenlinuxstb: I'm just wondering why i am not seeing them anywhere but in the files I added
18:07:37TheSevenat least not in the files I have currently opened
18:07:40linuxstbTheSeven: Programmer error? I've seen them in lots of places.
18:08:12kugelLlorean: that exception is limited to other code, not simply other sources
18:09:15kugelJdGordon usually forgets those, but he doesn't really mess in firmware/ often
18:09:33linuxstbTheSeven: So yes, if you see places where they are missing, it could be nice to add them.
18:09:59TheSeventhings like s5l8700.h for example
18:10:25linuxstbYes, I noticed that was missing it. Similar files have protection - e.g. pp5020.h
18:10:42TheSevenwhat's the convention for those define names in rockbox?
18:10:53TheSevenjust __BASENAME_H__?
18:10:55linuxstbIt would be _S5L8700_H
18:11:10 Join Utchybann [0] (
18:11:34funmanusb.h has an inclusion guard
18:11:56linuxstbSeems some files use BASENAME_H, some use _BASENAME_H and some __BASENAME_H
18:12:38linuxstbOh, and some use __BASENAME_H__
18:12:40kugelsome have __BASENAME_H__ too (mine :p)
18:12:49TheSeven...and mine
18:13:00linuxstbSo I guess the answer is there is no convention - do what you want ;)
18:13:08 Quit ender` (" It's all fun and games until someone loses an eye. Then it's fun and games you can't see.")
18:13:21 Join esperegu [0] (n=quassel@
18:13:26TheSevenanother thing that should be changed is the timeout handling in the nand driver - these loops are just evil
18:16:08funmangdb serial debugging is only implemented for archos ?
18:16:16 Quit Lss (Read error: 104 (Connection reset by peer))
18:16:29 Join Blue_Dude [0] (
18:17:15 Join Lss [0] (
18:18:33TheSevenhow do i get the element count of an array in C?
18:18:52TheSeven(in this case nand_deviceinfotable)
18:19:52funmanif the array is builtin constant you can use sizeof(array)/sizeof(array[0])
18:21:07TheSevengah, ugly. there really is no better way?
18:21:20polobricolois sound supported in the main rockbox ?
18:21:34Blue_DudeOk, quickscreen: how do I set up a menu so that a scrollable item works correctly? I can't get it to work for the compressor threshold.
18:21:34 Nick YPSY is now known as Ypsy (
18:21:35kugelwhich is just the wrapper for what funman wrote
18:21:38polobricoloor you add an empty element at the end
18:22:02polobricolois sound supported in the main rockbox *for nano2g
18:22:07funmanpolobricolo: no
18:22:29polobricologames work :)
18:22:40TheSevenpolobricolo: regarding sound: there is some code, and there are some bugs
18:22:52kugelfunman: that only works for builtin constants?
18:23:05funmankugel: yes sizeof() can't guess dynamic array size
18:23:20funmanbut dynamic memory allocation is a rare thing in rockbox no ? ;)
18:23:26TheSevenwell, are there such things as dynamic arrays in rockbox, without malloc?
18:23:48kugelfunman: there's also dynamic arrays on stack
18:24:17YpsyMy Rockbox keeps crashing when trying to start music (dir 3 is not free) and when I connect it to the PC its not writable anymore :/
18:24:20polobricoloi remember seeing on the rockbox status page sound only worked on bootloader, but i was wondering if TheSeven's patch added it to the main rockbox
18:24:25funmankugel: then you know the size, let me check what sizeof() does there
18:24:45YpsyThink it hapenned after doom crashed... I've got a sansa fuze and the newest versions
18:24:53YpsyJust installed it yesterday
18:24:56linuxstbYpsy: What device are you running Rockbox on? It sounds like the filesystem is broken - run something like chkdsk or fsck.vfat
18:25:36Ypsyfrom linux onto the player? or has rockbox an option for that?
18:25:49linuxstbYes, from Linux.
18:26:04Ypsycould I also just format it? or what that possibly brick it?
18:26:27Ypsywould that* i mean
18:27:54TheSevenpolobricolo: i didn't do anything with sound in rockbox. it's currently broken even in the bootloader.
18:28:08 Join ender` [0] (
18:28:16linuxstbTheSeven: Seems "SECTOR_SIZE" is used elsewhere in firmware/
18:28:30Ypsywell I'll just try to chkdsk it :) ty! laters
18:28:37funmankugel: sizeof() seems to work for stack allocated arrays as well
18:28:53 Join Guest23293 [0] (
18:29:04 Nick Ypsy is now known as YPSY (
18:29:06TheSevenlinuxstb: what are you talking about? USB storage?
18:29:13linuxstbVarious places - use grep
18:29:51Blue_DudeAnybody, need help with quickscreen. I can't get it to behave with my menu...
18:29:55linuxstbSome drivers seem to be using it. I guess those definitions could just be removed (I think they are all just defining it to 512), but it makes the patch riskier...
18:30:25TheSevenwhat the hell.
18:31:54TheSeventhese seem to be all storage drivers, so i don't see the problem as none of them are nano-related
18:32:03TheSevenas gevaerts said, this should be consolidated somehow
18:32:34TheSevenbut the current solution can't break anything, it could just be that defining a different sector size for other targets may or may not work
18:33:00*linuxstb sees it defined in ramdisk.c, ata.c, ata_flash.c, usb_storage.c, ata-nand-telechips.c, ata-sd-pp.c, ata-creativezvm.c, ata_sd_as3525.c
18:33:18linuxstbI guess the issue is that we can potentially have devices with multiple disks, each with different sector sizes.
18:34:01TheSevenwhich header is that ARRAYLEN declared in?
18:34:14kugelsystem.h IIRC
18:35:08TheSevenlinuxstb: Yes, I know that we can't move some targets to other sector sizes right now without breaking something, but this patch is a first step to allow cleaning this up later
18:35:36 Part pondlife
18:36:30TheSevenlinuxstb: should i split this patch again, or will you commit it as one huge commit anyways?
18:37:15linuxstbSplitting it is useful - I prefer to commit in small parts, so the commit message is more useful.
18:38:04BagderFYI: I'm off on vacation for two weeks starting in a few hours
18:38:26saratogai don't mind mixed case defines as long as they have a lot of capitals, rockbox variables are almost always lowercase so anything with capitals looks like a macro to me
18:38:31 Join Strife1989 [0] (
18:38:33 Quit Strife89 (Read error: 104 (Connection reset by peer))
18:39:35kugelBagder: again?
18:39:43linuxstbTheSeven: I'm just about to commit patch #1 - so no need to repost that.
18:40:08TheSevenyep, i only change #2 and #5
18:40:39kugelnot your first vacations this year IIRC (anyway, it doesn't matter and is off-topic)
18:41:35saratogafunman: interesting result in my clip test thread
18:41:50saratogathe first bug report is for the case in the FS task that we thought couldn't actually happen
18:42:05CIA-43New commit by dave (r22874): Patch #1 from FS #10633 (Nano 2G developments) by Michael Sparmann - Allows targets to set a different sector size than 512 for the storage system. ...
18:42:27 Part toffe82
18:42:39funmansaratoga: which thread ?
18:42:59linuxstbTheSeven: Sorry, I need to run now, but will look at your other patches later this evening.
18:43:57saratogafunman: it would seem both parts of FS #10605 are useful
18:44:13saratogaalthough from my reading of the code neither should really be able to happen
18:44:25 Quit cool_walking_ (Read error: 54 (Connection reset by peer))
18:45:53funmansaratoga: which case couldn't happen ? (i don't remember irc discussion, and fs#10605 has not much details)
18:46:02saratogawell niether really
18:46:11 Quit n17ikh (No buffer space available)
18:46:15saratogait shouldn't be possible to have to prebuffer the mp3 codec if you only play mp3
18:46:36saratogaand it shouldn't be possible for the overlap between a buffer handle position to be larger then the buffer handle
18:46:43saratogaat least as far as I can tell
18:46:54saratogaand yet I have logs documenting deadlocks related to both
18:47:12funmanah I see the case in apps/buffering.c diff
18:47:35saratogaIMO we should disable prebuffering codecs on small memory targets since its obviously broken and serves no productive purpose even if it worked
18:47:42TheSevennew patches in FS #10633
18:47:44funmanyou should have logged overlap & size_to_move to see if they are meaningful
18:47:57saratogai'll change that now and put up a new build
18:48:06saratogai wasn't able to get that case to happen so i assumed it couldn't
18:49:40funmansaratoga: i can test clip builds, but i won't invest much time in clip development atm
18:50:21 Join webguest62 [0] (i=www-data@
18:50:22 Quit daurn (Read error: 110 (Connection timed out))
18:50:30saratogathats fine
18:50:50saratogai'm not sure more time is really needed, just testing to see whats actually going on in playback
18:51:02saratogawell some of Nico's time woudl be nice
18:52:28linuxstbTheSeven: I'll do an s/crap/incorrect/ in system-s5l8700.c
18:53:11webguest62Hi... sorry to interrupt.. I just heard of rockbox and was wondering if it worked on a philips GoGear sa3020/37? am I in the Right channel?
18:53:28CIA-43New commit by dave (r22875): Patch #2 from FS #10633 (Nano 2G developments) by Michael Sparmann - Some hardware define corrections for the S5L8701, needed for the FTL.
18:53:34kugelsaratoga: my log
18:53:47TheSevenlinuxstb: I'm fine with that
18:53:58 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
18:54:02kugelI don't know if playback froze at some point, since the clip deadlocked completely after an hour
18:54:06linuxstbwebguest62: Yes, you're in the right channel, but I think the answer is no. See for information about the devices Rockbox runs on.
18:54:37webguest62Okay... I didn't see anything though. thank you!
18:54:54TheSevenwebguest62: if it is supported, you'll see it.
18:55:02webguest62Okay bye!
18:55:08 Quit webguest62 (Client Quit)
18:55:32funmanah, my yh920 sees a data abort on the COP
18:56:10funmannow in get_non_intra_block()
18:57:08saratogakugel: hmm ok, not sure what to make of that one
18:57:41funmani don't see defines about shared data in config-*.h
18:57:57linuxstbTheSeven: Is #3 actually needed for the flash driver? Or is it just bootloader and ibugger related?
18:58:20Blue_Dudewebguest62: is it listed on the main page?
18:58:36saratogatoo late!
18:58:47linuxstbTheSeven: But I'm really leaving now - back later.
18:59:05 Quit linuxstb ("Leaving")
18:59:12kugelsaratoga: it reloads the codec a lot after line 70
18:59:29Blue_DudeMy external HDD sucks! Drops posts all the time. I hate it.
18:59:40kugelthe end looks suspicious too
18:59:49saratogakugel: yeah but it eventually recovers and plays a lot longer without doing that
19:00:10saratogareset_handle(57) is really odd but i've never seen it before
19:00:13kugelthe end may come from the reboot after the freeze though
19:01:05Blue_DudeFrak, I'm going to swap this thing out. Back later...
19:01:05 Quit Blue_Dude ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
19:01:10kugelit resets the same handle thrice?
19:01:16saratogaaudio: initializing is probably the first line from after the reboot
19:01:40saratogawhen I shut down cleanly there is no end of log message, though I should probably add one
19:02:41 Join bertrik [0] (
19:03:11saratogathat log is wierd since it doesn't hit any of the cases we know to be problematic
19:03:37saratogai guess what we should do is get a few more logs with more or less SVN buffering, then start working around the cases that we know about and see what gets reported then
19:05:33funmanbertrik: how's playback stability on c200v2 ? did you try the clip patch ?
19:06:07bertrikplayback stability with SVN on the c200v2 is very poor
19:06:26funmanbertrik: have a look at the forum thread saratoga made for clipv1, there's a patch about logging
19:06:30***Saving seen data "./dancer.seen"
19:06:43saratogaprobably due to lack of RAM, c200v2 should have even less then clip
19:07:20funmanalbum art is problematic, i had a patch to prevent loading too big covers indefinitely but i was not sure about it
19:07:40bertrikwith the buffering_flash patch it was a little better (plus AA disabled)
19:08:22bertriki.e. able to play a full song but still crashing occasionally
19:08:47saratogabertrik: try the first patch here:
19:09:09saratogait helps a lot with the sim when the buffer is shrunk below that of the clip
19:09:33bertriksaratoga, does the sim help to narrow down the problem, or do we need to run with the actual target?
19:10:00bertriknever mind
19:10:50*bertrik should keep reading IRC when typing a question
19:15:00bertrikI think mc2739 has a c200v2 too
19:16:15 Join stoffel [0] (
19:16:18funmanbertrik: that's not an excuse :P
19:18:28TheSevenlinuxstb (for the logs): I think the crt0.S in SVN is buggy, and a part of that patch is also in your code base, IIRC.
19:19:00TheSevenso i would go for committing it - or maybe having some meizu tester (bertrik?) test it before
19:19:46TheSevenworst case is s5l8700 breakage, so I consider it less risky than the clickwheel patch, which *may* affect "production targets"
19:20:16TheSevenbertrik: do you have time to check if a nano2g patch modifying crt0.S has any adverse effects on meizu?
19:20:16bertrikmeizu doesn't really do much, but I won't mind testing to see if the bootloader demo still runs
19:20:21 Quit robin0800 (Read error: 110 (Connection timed out))
19:20:51TheSevenbertrik: then please try patch #3 from FS
19:20:56TheSevenFS #10633
19:21:33bertrikfunman, saratoga playback on the c200v2 still runs after a few minutes of playback/skipping/seeking (with the first patch of 10605)
19:25:04 Join aevin [0] (i=eivindsy@unaffiliated/aevin)
19:25:43 Quit daurn (Read error: 113 (No route to host))
19:28:29 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
19:28:35TheSevenmarkun: please think of better reg names for some things you put into s5l8700.h
19:28:44TheSeventhis one is causing dozens of define clashes
19:29:01TheSeven2 or 3 letter names for regs isn't really appropriate
19:29:26bertrikI think markun copied those from the data sheet
19:30:29bertrikugh, current SVN no longer runs on the meizu m3
19:30:57funmanit looks like a gdb stub is only written for ifp
19:31:00funmanand archoses
19:32:08*bertrik tries a clean tree
19:33:34TheSevenbertrik: i have modified some setup code in crt0.S, and it looks like a bug slipped in: try adding some code to re-open all the clock gates, that seems to have got lost.
19:33:59TheSevendon't think the protection unit stuff will really affect the meizu build
19:34:52bertrikoh, a build from a clean tree still works, now see what the patch does
19:35:50TheSeveni basically rewrote the protection unit and caching setup - and accidentally killed the clock gate thing
19:35:59TheSeven(and some other nano2g-ifdef'ed changes)
19:36:18TheSevenhow do I install a rockbox build i have just compiled? i.e. all the plugins and codecs?
19:37:04funmanmake zip, unzip at player's root
19:37:34bertrikTheSeven, it no longer works ...
19:38:44TheSevencan you try to just patch the actual protection unit setup code, to see if this is the issue?
19:38:52 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
19:39:19TheSevenhey, rockbox looks nice now, with a backdrop, icons and stuff :-)
19:40:07 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:40:08 Quit polobricolo (Remote closed the connection)
19:40:48bertriksaratoga, funman , still playing ...
19:41:10linuxstbTheSeven: You probably want to do a "make fullzip", to install the fonts as well (unless you already have those from an earlier install)
19:41:10TheSevenlinuxstb: #3 indeed seems to break meizu for some reason
19:41:12 Join polobricolo [0] (
19:42:05TheSevenfonts didn't look bad even though i didn't have them on there yet
19:42:22kugelTheSeven: picture or didn't happen!
19:42:22linuxstbYes, the system font is OK on small LCDs.
19:42:39TheSevenkugel: let me first fix the fonts
19:42:52funmanmpegplayer tries to access yh920 memory at 0xAD3F0E8F
19:42:58kugelmake zip should give you some fonts, the one for cabbiev2 for sure
19:43:21linuxstbAh yes - I forgot that.
19:43:33 Join Thundercloud [0] (
19:43:55funmani'm still unsure how lcd code does affect mpegplayer :/
19:44:26kugelmpegplayer calls the blit functions? that's all I know :p
19:44:31TheSevenlinuxstb: the plugins menu is broken
19:44:39TheSeveni can start plugins, but their names aren't shown
19:44:47TheSevenso it's a little poking in the dark :-P
19:44:53kugeldoes it lack some alignment maybe? several targets align the pixels to be drawn
19:45:04funmancube plugin works fine (except i can't leave it ..)
19:45:35kugelah greylib, doesn't it have its own target-independant blit ?
19:45:35linuxstbTheSeven: The plugins menu is just a file browser - so that would point to the nand driver not working. Does the normal file browser work if you browse to .rockbox/rocks/ ?
19:45:45TheSevenyes, it does
19:45:51TheSevenand the plugins get started and work properly
19:45:52funmanmandelbrot as well
19:45:58TheSevenit's just that the names are missing in the menu
19:46:04linuxstbThat's odd.
19:46:48TheSeven"plasma.rock" is having vsync issues - you see a stripe wandering around
19:47:19linuxstbTheSeven: I need to run again... Back again later.
19:47:49bertrikI don't think we vsync at all in rockbox
19:49:11TheSevenis there any generic key to leave plugins?
19:50:07funmanit's device specific
19:50:34TheSevenrunning pictureflow yields a stack overflow
19:51:08CIA-43New commit by funman (r22876): Samsung YH920 : greylib support based on ipod/lcd-as-gray.S ...
19:52:29UtchybannTheSeven: generally menu+select works in plugins
19:52:44TheSevenok, it starts fine if i increase stack size to 16KB
19:52:51UtchybannTheSeven: on my ipod mini.
19:55:48TheSevenok, FTL seems to be rock solid until now
19:56:32TheSevennow audio, huh?
19:57:09 Nick Guest23293 is now known as n17ikh (
19:58:20 Join dfkt_ [0] (i=dfkt@unaffiliated/dfkt)
19:59:59 Quit esperegu (Read error: 113 (No route to host))
20:00:19TheSevenbertrik: did that patch break meizu or not? I'm a little confused.
20:00:41TheSevenand what's the recommended way to test the audio stuff?
20:01:16TheSevenand how do i cleanly reboot rockbox? (and where was that place to put shut shutdown code to?)
20:03:43funmanTheSeven: look for power_off() and system_reboot()
20:04:05TheSevenhow do i trigger such a shutdown as the user?
20:04:35funmanhold POWEROFF_BUTTON
20:04:40 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
20:06:54TheSevenwhatever this is.
20:07:11funmanit should be in your button-target.h
20:08:03bertrikTheSeven, yes, the 3_... patch prevents the meizu m3 from booting properly
20:08:45amiconnTheSeven: Stack overflow is very likely due to the increased SECTOR_SIZE
20:09:41TheSevenamiconn: yep, that's suing up to 4KB of stack during read accesses
20:10:22TheSevenbertrik: any idea which part of it is causing that?
20:10:37TheSevenhave you tried only applying the protection unit setup code?
20:10:44 Join Guest23293 [0] (
20:10:53bertrikno, not yet
20:11:12 Quit n17ikh (Nick collision from services.)
20:11:14 Nick Guest23293 is now known as n17ikh (
20:11:53TheSevenfunman: and how do i trigger a reboot?
20:12:30UtchybannTheSeven: I only know menu+select (as OFW).
20:12:47funmanTheSeven: i don't think there is a shortcut for reboot (it's used in exception handler however)
20:14:31TheSevenok, I'm making power_off() reboot then, as we don't know how to power this thing off properly anyways
20:15:28funmantry throwing it against walls ;)
20:16:24 Quit dfkt (Read error: 110 (Connection timed out))
20:16:35AlexPUtchybann: That's a "hard" reboot
20:21:24TheSevenwhich header do i need to include to use splashf?
20:22:32*funman gives grep -r splashf firmware/export to TheSeven
20:23:39TheSevenfunman: it isn't there
20:24:07TheSevenapps/gui/splash.h :-/
20:24:08funmanah it's in apps/ only
20:27:12 Join esperegu [0] (n=quassel@
20:27:46 Join moonscapex [0] (
20:29:09 Quit moonscapex (Client Quit)
20:31:01 Quit Lynx_ (" Want to be different? Try HydraIRC -> <-")
20:31:28amiconnTheSeven: We're not doing vsync in rockbox because it is impossible on almost all targets
20:31:30saratogabertrik: on the SIM with less then 200KB buffer, playback locks up pretty quickly due to frequent (and erroneous) attempts to prebuffer the next tracks codec
20:31:57saratogathis still happens on targets with 8MB of memory, although since there is so much buffer available it rarely fails, just wastes some RAM
20:32:13saratogawith 64MB i didn't observe it happening
20:32:35amiconnThose using serial connection to the lcd controller are usually write-only, and those where the lcd controller uses parallel connection most of the time uses a controller that doesn't allow reading its current position
20:33:54TheSevenamiconn: I'm wondering if I'm actually seeing vsync issues or rather blitting in the middle of drawing the next frame
20:34:19funmanamiconn: any idea why making lcd_blit_* write to the LCD crashes mpegplayer? commenting one instruction causes a crash at a different address so i was supposing different alignements
20:34:47amiconnThis happens in mpegplayer only, right?
20:34:53funmanI tried increasing CACHEALIGN_SIZE to 32 bytes instead of 16, but I suppose caching is common to all PP
20:34:57funmanyes, on the COP
20:34:58amiconnAnd your target is PP, i.e. dualcore?
20:35:14funman(data abort on current_core == 1)
20:35:37amiconnI suspect that something is done differently in the YH920 lcd driver, which will hit in such a case
20:36:00funmanyh920 lcd is based on coldfire h100, i'll check ipod lcd code
20:36:47amiconnBtw, it's not only mpegplayer which puts the greyscale isr on the cop, but iirc it's the only plugin that also uses the cop for other things as well
20:37:18TheSevenwhat's the header to include to have memcpy, memset, etc.?
20:37:31funmanstring.h ?
20:37:31bertrik<string.h> I suppose
20:38:02 Quit Strife1989 (Read error: 104 (Connection reset by peer))
20:38:10TheSevennot really straight-forward naming
20:38:26 Join Strife1989 [0] (
20:38:53kugelsaratoga: I made another run. the clip shut off at some time this time, but the log looks totally fine (reusing the codec always, none of your !!! or @@@ marks)
20:39:21amiconnstring.h is very standard
20:39:24funmanamiconn: where is greyscale isr defined, and set to run on COP ?
20:39:29kugelTheSeven: blame the standard committee :)
20:40:36amiconnfunman: The greylib init call specifies whether the isr should run on cop. However, this works on all portalplayers I was able to test on, so the bug must be in target specific code
20:41:26funmanamiconn: sure, i suppose some shared data isn't properly accessed from uncached memory, so i want to see exactly which functions run on the COP
20:41:31amiconnIt *could* be in lcd format specific code in the greylib as well, as dualcore is only tested on horizontally packed 2bpp so far (all other supported packings were non-PP)
20:41:48 Join nls_web [0] (
20:42:38 Quit Rondom (Nick collision from services.)
20:42:48 Join Rondom [0] (
20:45:58*TheSeven found another FTL bug, which prevented it from unmounting cleanly
20:46:24amiconnWhy is unmounting needed?
20:46:49TheSevenbecause the bootloader will need to do a consistency check on the next boot
20:46:57*amiconn thinks that a battery powered embedded device shouldn't do any long-term write caching
20:47:08TheSevenit's not about caching, but wear reduction
20:47:23TheSevenyou don't always want to erase a 1MB block of the flash because someone wrote 2KB
20:48:37TheSeven(the OFW bootloader seems to clean this up on the next reboot, but this takes ~30sec)
20:48:45amiconnHow do memory cards handle that? I know that if I write a 512 byte block to an mmc, it is guaranteed to be written to the flash if it signals write completion. An (internal) erase block is certainly much larger than 512 bytes
20:49:22TheSeventhe data is written to flash here, too, but not to it's final location, but to something called scattered page block
20:49:59TheSevenand if there is an unclean shutdown, the bootloader has to scan a huge amount of space to collect these leftovers and properly commit them
20:51:09TheSevenso an unclean shutdown will probably not mean data loss, but a nasty delay when mounting the FTL the next time
20:52:22 Quit bubsy (Read error: 60 (Operation timed out))
20:53:08 Join Blue_Dude [0] (
20:53:44Blue_DudeHaving problems with SVN update. Files transfer OK, but when the process ends, it hangs until I ctrl-C. I'm using cygwin.
20:56:37TheSevenI love such bugs. It *sometimes* unmounts cleanly.
20:56:59TheSevenalways, if the debuggers are in place, of course
21:00:03 Quit nls_web ("CGI:IRC")
21:00:39Blue_DudeAnyone else having issues with SVN?
21:01:59 Quit flydutch ("/* empty */")
21:02:23saratogaBlue_Dude: its working normally for me
21:02:41Blue_DudeRats. Hanging for me on multiple checkouts.
21:02:50saratogai would either check out a fresh svn copy, or else explore using the svn status or repair options (i forget what they are)
21:04:11 Join bubsy [0] (n=bubsy@
21:05:09Blue_Dudesvn status is OK. svn up hangs no matter which version I try to update.
21:06:34***Saving seen data "./dancer.seen"
21:06:37kugelcheckout in windows
21:06:43Blue_DudeI new version of subversion is available from cygwin. I'm going to install it first.
21:06:56kugelIIRC tortoisesvn can be configured to use unix line endings
21:07:11Blue_DudeDunno how to checkout in windows...
21:07:27Blue_Dudeoh ok
21:07:33kugeleither by using tortoisesvn or the cli client
21:07:34saratogayou should consider getting a linux vm
21:07:38saratogacygwin is painfully slow
21:07:39kugelthat too
21:07:47pixelmado you use tortoisesvn? It wasn't mentioned before
21:08:02Blue_DudeIs there anything that will run under windows?
21:08:26Blue_DudeI know 000.00 about linux.
21:08:50Blue_DudeNo, just cygwin.
21:08:58kugelI was in the same situation when I got into rockbox :)
21:09:36pixelmaBlue_Dude: I'm using plain subversion under cygwin too and an update just worked fine for me
21:09:38Blue_DudeI don't have the option of installing a native linux installation. It has to be something that runs on top of windows. Company laptop.
21:10:03kugeljust download virtualbox and install ubuntu in a vm. that's the fail-safe first step into linux
21:11:16pixelmaBlue_Dude: did you try running "svn resolved" or hmm... forgot what the other was called, need to look it up
21:11:34Blue_DudeNo conflicts.
21:12:49*kugel still regrets having tried cygwin at all
21:12:56 Join nls_web [0] (
21:13:04kugelBlue_Dude: sometimes svn just takes insanely long, try giving it some 5 minutes
21:13:46Blue_DudeI don't think I can use virtualbox. IT would freak.
21:14:41pixelmasvn cleanup I think it was. If I remember correctly I had to do it after a failed svn update once, but subversion told me that I had to
21:15:07 Quit funman ("free(random());")
21:17:19kugelBlue_Dude: why?
21:18:51 Quit pcc1 (
21:18:51 Quit r00s (
21:18:51 Quit ps-auxw (
21:18:51 Quit markun (
21:18:51 Quit bzed (
21:18:51 Quit parafin (
21:18:51 Quit n17ikh (
21:18:51 Quit FlynDice (
21:18:51 Quit maraz_ (
21:18:51 Quit mc2739 (
21:18:51 Quit sbhsu (
21:18:51 Quit Bagder (
21:18:51 Quit GodEater (
21:18:51 Quit scorche (
21:18:51 Quit Blue_Dude (
21:18:51 Quit Rondom (
21:18:51 Quit Strife1989 (
21:18:51 Quit polobricolo (
21:18:51 Quit DerPapst (
21:18:51 Quit freqmod (
21:18:51 Quit Trista887 (
21:18:51 Quit Llorean (
21:18:51 Quit evilnick (
21:18:51 Quit saratoga (
21:18:51 Quit esperegu (
21:18:51 Quit linuxstb (
21:18:51 Quit crashd (
21:18:51 Quit niekie (
21:18:52 Quit dfkt_ (
21:18:52 Quit chandoo (
21:18:52 Quit Bob_C (
21:18:52 Quit tmzt (
21:18:52 Quit ch4os (
21:18:52 Quit TheSeven (
21:18:52 Quit hd (
21:18:52 Quit Rob2223 (
21:18:52 Quit T44 (
21:18:52 Quit togetic (
21:18:52 Quit Paprica (
21:18:52 Quit amiconn (
21:18:52 Quit rvvs89 (
21:18:52 Quit fxb (
21:18:52 Quit z35 (
21:18:52 Quit ChanServ (
21:18:52 Quit daurn (
21:18:52 Quit Lss (
21:18:52 Quit preglow (
21:18:52 Quit cg_ (
21:18:52 Quit gtkspert_ (
21:18:52 Quit killan_ (
21:18:52 Quit advcomp2019_ (
21:18:52 Quit kadoban (
21:18:52 Quit Dhraakellian (
21:18:52 Quit linuxguy3 (
21:18:52 Quit pjm0616 (
21:18:52 Quit jfc (
21:18:52 Quit jordan` (
21:18:57 Quit tarbo (
21:18:57 Quit jvd (
21:18:57 Quit aevin (
21:18:57 Quit Hadaka (
21:18:57 Quit AB3JU (
21:18:57 Quit Galois (
21:18:57 Quit meermanr (
21:18:57 Quit mikroflops (
21:18:57 Quit rwong (
21:18:57 Quit J-23 (
21:18:57 Quit Tuplanolla (
21:18:57 Quit lyngaas (
21:18:57 Quit DaveDavenport (
21:18:57 Quit ender` (
21:18:57 Quit _lifeless (
21:18:57 Quit JackWinter (
21:18:57 Quit gevaerts (
21:18:57 Quit lostlogic (
21:18:57 Quit sinthetek (
21:18:57 Quit krazykit (
21:18:57 Quit AlexP (
21:18:57 Quit Beta2K (
21:18:57 Quit DirtySkeMe (
21:18:57 Quit avacore (
21:18:57 Quit yosafbridge (
21:18:57 Quit rphillips (
21:18:57 Quit DaCapn (
21:18:57 Quit zu (
21:18:57 Quit CIA-43 (
21:18:57 Quit bubsy (
21:18:57 Quit stoffel (
21:18:57 Quit Utchybann (
21:18:57 Quit goffa (
21:18:57 Quit rasher (
21:18:57 Quit ThomasAH (
21:18:57 Quit liar|netbook (
21:18:57 Quit thegeek (
21:18:57 Quit panni_ (
21:18:57 Quit bmbl (
21:18:57 Quit jon-kha (
21:18:57 Quit Slasheri (
21:18:57 Quit w1ll14m|_ (
21:18:57 Quit Torne (
21:18:57 Quit Shaid (
21:18:57 Quit elcan (
21:18:57 Quit dionoea (
21:18:57 Quit KBH (
21:18:57 Quit tha (
21:18:57 Quit blithe (
21:18:57 Quit janne_oksanen (
21:18:57 Quit scorche|sh (
21:18:57 Quit alexbobp (
21:19:16NJoinChanServ [0] (ChanServ@services.)
21:19:16NJoinbubsy [0] (n=bubsy@
21:19:16NJoinBlue_Dude [0] (
21:19:16NJoinRondom [0] (
21:19:16NJoinStrife1989 [0] (
21:19:16NJoinesperegu [0] (n=quassel@
21:19:16NJoinn17ikh [0] (
21:19:16NJoindfkt_ [0] (i=dfkt@unaffiliated/dfkt)
21:19:16NJoinpolobricolo [0] (
21:19:16NJoinlinuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
21:19:16NJoindaurn [0] (i=daurnima@freenode/staff/daurnimator)
21:19:16NJoinaevin [0] (i=eivindsy@unaffiliated/aevin)
21:19:16NJoinstoffel [0] (
21:19:16NJoinender` [0] (
21:19:16NJoinLss [0] (
21:19:16NJoinUtchybann [0] (
21:19:16NJoin_lifeless [0] (n=lifeless@
21:19:16NJoinchandoo [0] (
21:19:16NJoinDerPapst [0] (
21:19:16NJoinpreglow [0] (i=thomj@rockbox/developer/preglow)
21:19:16NJoinliar|netbook [0] (n=liar@
21:19:16NJoinevilnick [0] (i=0c140464@gateway/web/freenode/x-vssmtmmrwwvlkrzw)
21:19:16NJointhegeek [0] (n=nnscript@
21:19:16NJoinpanni_ [0] (i=hannes@
21:19:16NJoinfreqmod [0] (
21:19:16NJoinbmbl [0] (n=Miranda@unaffiliated/bmbl)
21:19:16NJoincg_ [0] (
21:19:16NJoinBob_C [0] (
21:19:16NJointmzt [0] (
21:19:16NJoinch4os [0] (n=ch4os@gentoo/user/ch4os)
21:19:16NJoinTrista887 [0] (
21:19:16NJoinTheSeven [0] (
21:19:16 Join hd [0] (i=jd@Wikipedia/HellDragon)
21:19:16NJoinJackWinter [0] (
21:19:16NJoinRob2223 [0] (
21:19:16NJoinT44 [0] (
21:19:16NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
21:19:16NJointogetic [0] (n=togetic@unaffiliated/ibuffy)
21:19:16NJoinLlorean [0] (n=DarkkOne@rockbox/user/Llorean)
21:19:16NJoinamiconn [0] (i=quassel@rockbox/developer/amiconn)
21:19:16NJoinKBH [0] (n=hbk@
21:19:16NJoinz35 [0] (
21:19:16NJoinAlexP [0] (n=alex@rockbox/staff/AlexP)
21:19:16NJoinPaprica [0] (
21:19:16NJoincrashd [0] (
21:19:16NJoinscorche [50] (n=scorche@rockbox/administrator/scorche)
21:19:16NJoinbzed [0] (
21:19:16NJoinadvcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
21:19:16NJoinsaratoga [0] (i=9803c6dd@gateway/web/freenode/x-lklytxvqbpsofbwg)
21:19:16NJoinfxb [0] (
21:19:16NJoinrasher [50] (n=rasher@rockbox/developer/rasher)
21:19:16Mode"#rockbox +o ChanServ " by
21:19:16NJoinpcc1 [0] (
21:19:16NJoinmarkun [50] (n=markun@rockbox/developer/markun)
21:19:16NJoinps-auxw [0] (
21:19:16NJoinr00s [0] (
21:19:16NJoinmikroflops [0] (
21:19:16NJoinrvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
21:19:16NJoinmc2739 [0] (
21:19:16NJoinkillan_ [0] (
21:19:16NJoinmaraz_ [0] (
21:19:16NJoinGodEater [0] (n=bibble@rockbox/staff/GodEater)
21:19:16NJoinsbhsu [0] (
21:19:16NJoinBagder [241] (n=daniel@rockbox/developer/bagder)
21:19:16NJoinFlynDice [0] (
21:19:16NJoinparafin [0] (
21:19:16NJoinavacore [0] (
21:19:16 Join DaveDavenport [0] (n=qball@unaffiliated/qball)
21:19:16NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
21:19:16NJoingtkspert_ [0] (
21:19:16NJoinlinuxguy3 [0] (
21:19:16NJoinShaid [0] (i=adam@
21:19:16NJoinkrazykit [0] (
21:19:16NJoinTorne [0] (i=torne@
21:19:16NJoinjvd [0] (n=syscrash@poipu/developer/syscrash)
21:19:16NJoinBeta2K [0] (
21:19:16NJoingoffa [0] (n=goffa@
21:19:16NJoinniekie [0] (
21:19:16NJoinDirtySkeMe [0] (
21:19:16NJoinjfc [0] (
21:19:16NJoinalexbobp [0] (n=alex@
21:19:16NJoinjanne_oksanen [0] (n=janne@
21:19:16NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
21:19:16NJoinrphillips [0] (
21:19:16NJoinyosafbridge [0] (
21:19:16NJoinGalois [0] (
21:19:16NJoinThomasAH [0] (
21:19:16NJoinw1ll14m|_ [0] (n=nobody@
21:19:16NJoinpjm0616 [0] (n=user@
21:19:16NJointarbo [0] (n=me@unaffiliated/tarbo)
21:19:16NJointha [0] (i=1038@
21:19:16NJoinblithe [0] (n=blithe@
21:19:16NJoinjon-kha [0] (i=jon-kha@
21:19:16NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
21:19:16NJoinzu [0] (
21:19:16NJoinCIA-43 [0] (n=CIA@
21:19:16NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
21:19:16NJoinsinthetek [0] (
21:19:16NJoinelcan [0] (i=user36@
21:19:16NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
21:19:16NJoinJ-23 [0] (
21:19:16NJoinmeermanr [0] (
21:19:16NJoinrwong [0] (
21:19:16NJoinDaCapn [0] (
21:19:16NJoinlyngaas [0] (
21:19:16NJoinAB3JU [0] (
21:19:16NJoinHadaka [0] (
21:19:16NJoinDhraakellian [0] (
21:19:16NJoinkadoban [0] (
21:19:16NJoinjordan` [0] (i=gromit@
21:20:44 Quit nls_web ("CGI:IRC (EOF)")
21:21:37Blue_Dudekugel: "IT" would freak. Company IT dept.
21:22:07kugelyea, why would it freak?
21:23:12pixelmawhy would they freak
21:23:23Blue_DudeThey have an insanely strict installation policy. I'm probably not supposed to run cygwin either, but I get away with it because it doesn't screw with the registry and it's on my own HDD. If virtualbox had a portable version I'd be OK.
21:23:35 Quit rvvs89 (Remote closed the connection)
21:24:26 Quit stoffel (Remote closed the connection)
21:24:42kugelBlue_Dude: google "virtualbox portable"...
21:25:32kugelyou could also try qemu which is a more lightweight virtualization tool
21:26:17AlexPBagder: Could you add the e200v2 manual to the daily manual build, and to the manual page?
21:26:39TheSevenwriting still isn't as stable as I had hoped...
21:27:04 Quit bmbl (Connection timed out)
21:27:32kugelAlexP: badger is gone for 2 weeks
21:27:48AlexPkugel: I'm talking to him now
21:27:59AlexPBut I'm going to bug Zagor instead
21:28:07kugeland I think those pages are in svn (www/), someone needs a manual update
21:28:07CtcpVersion from freenode-connect!freenode@freenode/bot/connect
21:28:09TheSevenin fact, it seems to work mostly as long as you don't unmount. unmounting sometimes brings back old data
21:29:07AlexPkugel: I can't add the manual to the server build round, even if I can edit the page
21:29:10 Join rvvs89_ [0] (n=ivo@
21:29:19kugelAlexP: oh I didn't look at -community, nevermind then
21:34:26 Join rvvs89 [0] (n=ivo@
21:36:11TheSevenI hate bugs that get scared from splashf.
21:38:10 Quit rvvs89_ (Read error: 145 (Connection timed out))
21:39:37 Join rvvs89_ [0] (
21:40:33 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
21:41:33 Join froggyman [0] (
21:41:41*TheSeven suspects some kind of memory corruption could be the cause
21:43:54 Quit rvvs89 (Read error: 145 (Connection timed out))
21:44:42 Quit rvvs89_ (Remote closed the connection)
21:44:50 Join rvvs89 [0] (
21:47:22saratogafor some reason i never get any deadlocks on my clip using this logging build
21:47:29saratogaits been going for hours
21:49:58 Quit kkurbjun ("Leaving.")
21:53:49saratogahuh just had a real crash
21:53:54saratogait even rebooted the clip for me
21:54:56 Part froggyman
21:55:22 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
21:56:50amiconnkugel: tortoisesvn can *not* be configured for unix line endings (unfortunately, and unlike tortoisecvs)
21:57:24 Quit rvvs89 (Remote closed the connection)
21:57:39 Join rvvs89 [0] (
21:57:49amiconnTheSeven: Shy bugs do that. They know you're looking at them ;)
21:58:31TheSevencan anybody tell me what IRAMSIZE in is supposed to be? does that include the IRAM used for codecs and plugins, or not?
21:58:50amiconnNo it doesn't
21:59:11amiconnIt's just the part reserved for the core
21:59:19TheSevenand if IRAM is full, i suppose I get a link time error?
21:59:35saratogayes if you overfill IRAM it won't link
22:00:16kugelsaratoga: the patch has FS #10605 in it
22:00:21TheSevenfrom looking at, it looks like codecs and plugins share the same iram space?
22:00:33 Join volkmar [0] (n=volkmar@gentoo/developer/volkmar)
22:00:37kugelI thought that removes all deadlocks
22:00:53amiconnTheSeven: Yes. That's the reason that plugins which use IRAM need to stop playback
22:01:01kugeloh, nevermind
22:01:09amiconn*reason why
22:01:47TheSeventhe behavior I'm currently getting depends massively on stack size
22:02:09saratogakugel: yeah its applied but disabled for now
22:02:13volkmarmy old iPod Nano (with Rockbox) has its lock switch broken so it's really hard to unlock it more than a second. Is it possible with a patched rockbox to do by-pass the lock switch ?
22:02:41 Quit rvvs89 (Remote closed the connection)
22:02:46 Join rvvs89 [0] (
22:02:52volkmarin other words, does someone know if the lock switch is locking the iPod directly or the software is getting the status change and refuse to get inputs
22:03:21amiconnIt depends on the target, but afaik it's hardware lock on the ipods
22:03:23TheSevendoes anyone have an explanation why *increasing* the stack size breaks the FTL?
22:03:57kugelare you doing clean makes?
22:04:24TheSevenam i supposed to do anything besides re-linking it?
22:04:28saratogamaybe the stack gets linked onto something important
22:04:29kugelI've had strange problems with touching linker-related stuff and not doing this (such as codecs don't work)
22:05:06TheSevenbut how can ever affect any compiler output?
22:05:06kugelif you tinker with the linker, rm -rf * is the safe way
22:06:16kugelnot relinking correctly (and silently) can cause all sorts of problems
22:07:19TheSevencat: /home/rockbox/rockbox/build/nano2-app/apps/features: No such file or directory
22:07:32TheSevenwhat the heck is going on with that working copy?
22:07:49 Quit rvvs89 (Remote closed the connection)
22:07:54 Join rvvs89 [0] (
22:08:08kugeltold ya :p
22:08:24TheSevenwell, that was after a "make clean"
22:09:18kugelrm -rf * != make clean. I did make clean also when I had the strange bugs, only rm -rf * would make them go away
22:10:18gevaertsthe only remaining files should be config.h and Makefile. I don't see what difference those can make
22:11:15saratogakugel: you did the charts on the SamsungRuntime page?
22:12:17kugelsaratoga: sure, they're mine
22:12:31saratogathe spike is probably when it switches from constant current to constant voltage charging
22:12:35 Join n1s [0] (n=n1s@rockbox/developer/n1s)
22:12:49kugelpossibly, I'm not very fit in that area
22:12:57 Quit rvvs89 (Remote closed the connection)
22:13:02 Join rvvs89 [0] (
22:13:14amiconngevaerts: 'make clean' won't delete everything if the location of some auto-generated changes
22:13:37saratogait looks normal enough to me, and since its a hardware charger it doesn't really matter to us anyway
22:14:48TheSevenhow much IRAM does the core usually need?
22:14:58amiconnAnd it's autoconf.h
22:15:05kugelgevaerts: don't ask me how, but when funman committed the working mp3 playback, all my codecs didn't run after svn up and doing a make
22:15:19kugelmake clean didn't help, but a deletion of the build dir did
22:16:00amiconnTheSeven: Changing just the stack size *shouldn't* require anything more than relinking. Check the .map for anything unusual
22:16:01saratogaTheSeven: it depends on teh target, but none have more then 48KB and most don't use all that
22:16:42TheSevenok, so a 96K (core) / 80K (plugins/codecs) split should be safe for a start?
22:17:25saratogayes thats more then enough unless you've put the frame buffer in IRAM or something similar
22:17:29amiconnLinking *should* fail if any section becomes over-full, but I've seen it missing this condition in fringe cases
22:18:07saratogaideally core should have just enough to work efficiently and the rest saved for codecs (which perform very well with 64KB but more is always helpful for increasing battery life)
22:19:09TheSevenwell, safety first, right now
22:19:36n1ssome of the greyscale targets are pretty close to the limit, like the h100 which iirc has both the frame buffers in iram
22:19:38kugelhow does increasing it help? doesn't one codec need X kB (statically)?
22:19:53TheSeveneverything pristine right now, killed the whole build directory, reinitialized the FTL
22:20:02TheSevenlet's see if the issues go away
22:20:14n1skugel: you can put more stuff in it if you make it larger?
22:20:20 Nick dfkt_ is now known as dfkt (i=dfkt@unaffiliated/dfkt)
22:20:29amiconnkugel: It only helps if you go through the codec and put some more thing into iram
22:21:19TheSevenwow. in fact, everything is fine now
22:21:34TheSeven(besides some other bugs, but the FTL stuff works at least)
22:22:00*kugel already said "told ya :p" so he'll be quiet this time
22:23:04saratogasome of the codecs have #ifdefs that check for extra IRAM
22:23:23TheSevensaratoga: huh? how do they know this?
22:23:25saratogaand of course if you save up more IRAM we can always find ways to use it
22:23:30TheSeveni thought that is defined in
22:23:38saratogaTheSeven: they just check the CPU model
22:23:50TheSevengah, hack.
22:23:52saratogaso they'd have to be updated for the nano2g but thats simple
22:24:22n1smaybe we should, uhm have some CODECIRAMSIZE define or something
22:24:24saratogawe typically optimize the codecs for new CPUs anyway so its not a big deal
22:24:35saratogathat'd be a good idea
22:24:37kugeladding new targets is a painful job for a number of reasons
22:24:46 Quit rvvs89 (Remote closed the connection)
22:24:51 Join rvvs89 [0] (
22:29:40*TheSeven would really like to get audio working
22:30:05TheSevenwhat are the steps required to do this?
22:38:15***Started Dancer V4.16
22:38:15***Connected to on port 6667
22:38:15***Logfile for #rockbox started
22:38:20***Server message 501: 'logbot :Unknown MODE flag'
22:38:20Mode"logbot :+i" by logbot
22:38:21 Join logbot [0] (
22:38:21 Join rvvs89 [0] (
22:38:21 Join n1s [0] (n=n1s@rockbox/developer/n1s)
22:38:21 Join volkmar [0] (n=volkmar@gentoo/developer/volkmar)
22:38:21 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
22:38:21 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
22:38:21 Join courtc [0] (n=court@unaffiliated/courtc)
22:38:21 Join crwl [0] (
22:38:21 Join Overand [0] (
22:38:21 Join shodanX [0] (
22:38:21 Join BlakeJohnson86 [0] (
22:38:21 Join timc [0] (n=aoeu@
22:38:21 Join redfox [0] (
22:38:21 Join SUSaiyan [0] (
22:38:21 Join bekabug [0] (
22:38:21 Join jds2001 [0] (n=jds2001@fedora/jds2001)
22:38:21 Join ej0rge [0] (
22:38:21 Join fish_ [0] (
22:38:21 Join YPSY [0] (
22:38:21 Join howell_ [0] (
22:38:21 Join Unhelpful [0] (n=quassel@rockbox/developer/Unhelpful)
22:38:21 Join Kohlrabi [0] (
22:38:21 Join shadearg [0] (
22:38:21 Join Zarggg_ [0] (
22:38:21 Join Res1 [0] (
22:38:21 Join FOAD [0] (
22:38:21 Join pixelma [0] (i=quassel@rockbox/staff/pixelma)
22:38:21 Join aidy [0] (
22:38:21 Join feisar-_ [0] (
22:38:21 Join Zambezi [0] (i=Zulu@
22:38:21 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
22:38:21 Join evilnick_home [0] (
22:38:21 Join antil33t [0] (n=Mudkips@
22:38:21 Join xavieran [0] (n=xavieran@
22:38:21 Join mt [0] (n=MTee@rockbox/developer/mt)
22:38:21 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
22:38:21 Join kugel [0] (n=kugel@rockbox/developer/kugel)
22:38:21 Join AndyI [0] (n=pasha_in@
22:38:21 Join faemir [0] (n=faemir@
22:38:21 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
22:38:21 Join bertrik [0] (
22:38:21 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
22:38:21 Join Thundercloud [0] (
22:38:21 Join bubsy [0] (n=bubsy@
22:38:21 Join Blue_Dude [0] (
22:38:21 Join Rondom [0] (
22:38:21 Join Strife1989 [0] (
22:38:21 Join esperegu [0] (n=quassel@
22:38:21 Join n17ikh [0] (
22:38:21 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
22:38:21 Join polobricolo [0] (
22:38:21 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
22:38:21 Join daurn [0] (i=daurnima@freenode/staff/daurnimator)
22:38:21 Join aevin [0] (i=eivindsy@unaffiliated/aevin)
22:38:21 Join ender` [0] (
22:38:21 Join Lss [0] (
22:38:21 Join Utchybann [0] (
22:38:21 Join _lifeless [0] (n=lifeless@
22:38:21 Join chandoo [0] (
22:38:21 Join DerPapst [0] (
22:38:21 Join preglow [0] (i=thomj@rockbox/developer/preglow)
22:38:21 Join liar|netbook [0] (n=liar@
22:38:21 Join evilnick [0] (i=0c140464@gateway/web/freenode/x-vssmtmmrwwvlkrzw)
22:38:21 Join thegeek [0] (n=nnscript@
22:38:21 Join panni_ [0] (i=hannes@
22:38:21 Join freqmod [0] (
22:38:21 Join cg_ [0] (
22:38:21 Join Bob_C [0] (
22:38:21 Join tmzt [0] (
22:38:21 Join ch4os [0] (n=ch4os@gentoo/user/ch4os)
22:38:21 Join Trista887 [0] (
22:38:21 Join TheSeven [0] (
22:38:21 Join hd [0] (i=jd@Wikipedia/HellDragon)
22:38:21 Join JackWinter [0] (
22:38:21 Join Rob2223 [0] (
22:38:21 Join T44 [0] (
22:38:21 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
22:38:21 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
22:38:21 Join Llorean [0] (n=DarkkOne@rockbox/user/Llorean)
22:38:21 Join amiconn [0] (i=quassel@rockbox/developer/amiconn)
22:38:21 Join KBH [0] (n=hbk@
22:38:21 Join z35 [0] (
22:38:21 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
22:38:21 Join Paprica [0] (
22:38:21 Join crashd [0] (
22:38:21 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
22:38:21 Join bzed [0] (
22:38:21 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
22:38:21 Join saratoga [0] (i=9803c6dd@gateway/web/freenode/x-lklytxvqbpsofbwg)
22:38:21 Join fxb [0] (
22:38:21 Join rasher [50] (n=rasher@rockbox/developer/rasher)
22:38:21 Join pcc1 [0] (
22:38:21 Join markun [50] (n=markun@rockbox/developer/markun)
22:38:21 Join ps-auxw [0] (
22:38:21 Join r00s [0] (
22:38:21 Join mikroflops [0] (
22:38:21 Join mc2739 [0] (
22:38:21 Join killan_ [0] (
22:38:21 Join maraz_ [0] (
22:38:21 Join GodEater [0] (n=bibble@rockbox/staff/GodEater)
22:38:21 Join sbhsu [0] (
22:38:21 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
22:38:21 Join FlynDice [0] (
22:38:21 Join parafin [0] (
22:38:21 Join avacore [0] (
22:38:21 Join DaveDavenport [0] (n=qball@unaffiliated/qball)
22:38:21 Join dionoea [0] (n=dionoea@videolan/developer/dionoea)
22:38:21 Join gtkspert_ [0] (
22:38:21 Join linuxguy3 [0] (
22:38:21 Join Shaid [0] (i=adam@
22:38:21 Join krazykit [0] (
22:38:21 Join Torne [0] (i=torne@
22:38:21 Join jvd [0] (n=syscrash@poipu/developer/syscrash)
22:38:21 Join Beta2K [0] (
22:38:21 Join goffa [0] (n=goffa@
22:38:21 Join niekie [0] (
22:38:21 Join DirtySkeMe [0] (
22:38:21 Join jfc [0] (
22:38:21 Join alexbobp [0] (n=alex@
22:38:21 Join janne_oksanen [0] (n=janne@
22:38:21 Join Tuplanolla [0] (n=jani@unaffiliated/tuplanolla)
22:38:21 Join rphillips [0] (
22:38:21 Join yosafbridge [0] (
22:38:21 Join Galois [0] (
22:38:21 Join ThomasAH [0] (
22:38:21 Join w1ll14m|_ [0] (n=nobody@
22:38:21 Join @ChanServ [0] (ChanServ@services.)
22:38:21 Join jordan` [0] (i=gromit@
22:38:21 Join kadoban [0] (
22:38:21 Join Dhraakellian [0] (
22:38:21 Join Hadaka [0] (
22:38:21 Join AB3JU [0] (
22:38:21 Join lyngaas [0] (
22:38:21 Join DaCapn [0] (
22:38:21 Join rwong [0] (
22:38:21 Join meermanr [0] (
22:38:21 Join J-23 [0] (
22:38:21 Join Slasheri [0] (i=miipekk@rockbox/developer/Slasheri)
22:38:21 Join elcan [0] (i=user36@
22:38:21 Join sinthetek [0] (
22:38:21 Join lostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
22:38:21 Join CIA-43 [0] (n=CIA@
22:38:21 Join zu [0] (
22:38:21 Join scorche|sh [50] (n=scorche@rockbox/administrator/scorche)
22:38:21 Join jon-kha [0] (i=jon-kha@
22:38:21 Join blithe [0] (n=blithe@
22:38:21 Join tha [0] (i=1038@
22:38:21 Join tarbo [0] (n=me@unaffiliated/tarbo)
22:38:21 Join pjm0616 [0] (n=user@
22:38:58amiconnsaratoga: Checking the cpu model is a bad idea. There are various IRAM_<feature> macros for this
22:39:52n1sTheSeven: the various pcm_* functions basically afair
22:40:20amiconnThese were originally introduced for the reverse case - one target (iFP7xx) having significantly less iram than most others.
22:46:35 Quit kugel (Nick collision from services.)
22:46:38 Join kugel [0] (
22:46:50 Quit kugel (Nick collision from services.)
22:46:57 Join kugel [0] (
22:47:05 Quit kugel (Nick collision from services.)
22:47:11 Join kugel [0] (
22:47:21 Quit kugel (Nick collision from services.)
22:47:27 Join kugel [0] (
22:47:39 Join kugel_ [0] (
22:47:50 Quit kugel (Nick collision from services.)
22:47:53 Nick kugel_ is now known as kugel (
22:48:05 Quit kugel (Nick collision from services.)
22:48:11 Join kugel [0] (
22:48:11 Join Omlet [0] (n=omlet05@
22:48:13 Join Omlet05 [0] (n=omlet05@
22:48:56 Join michaelcarr [0] (
22:49:06 Join Blue_Dude_ [0] (
22:49:07 Join kugel_ [0] (
22:49:12 Quit Strife1989 (Read error: 104 (Connection reset by peer))
22:49:18 Quit kugel (Nick collision from services.)
22:49:21 Nick kugel_ is now known as kugel (
22:49:33 Quit kugel (Nick collision from services.)
22:49:39 Join kugel [0] (
22:49:52 Join kugel_ [0] (
22:49:53*TheSeven wonders what's missing from the nano 2g audio implementation
22:50:00TheSevenif i look at that code, it seems to be mostly done
22:50:04 Quit kugel (Nick collision from services.)
22:50:08 Nick kugel_ is now known as kugel (
22:50:09 Quit Omlet (Client Quit)
22:50:20 Quit kugel (Nick collision from services.)
22:50:25 Join kugel [0] (
22:50:31TheSeven(borrowing wmcodec from some other device and i2s from the meizus)
22:50:35 Quit kugel (Nick collision from services.)
22:50:41 Join kugel [0] (
22:50:54 Join kugel_ [0] (
22:51:04 Quit kugel (Nick collision from services.)
22:51:08 Nick kugel_ is now known as kugel (
22:51:09TheSevenerm? kugel?
22:51:20 Quit kugel (Nick collision from services.)
22:51:26 Join kugel [0] (
22:51:35 Quit kugel (Nick collision from services.)
22:51:40 Join kugel [0] (
22:51:54 Join kugel_ [0] (
22:52:06 Quit kugel (Nick collision from services.)
22:52:10 Nick kugel_ is now known as kugel (
22:52:21 Quit kugel (Nick collision from services.)
22:52:22n1sTheSeven: i'd guess the most common thing is a mistake in the dac config or the i2s config
22:52:26 Join kugel [0] (
22:52:40 Join kugel_ [0] (
22:52:50 Quit kugel (Nick collision from services.)
22:52:54 Nick kugel_ is now known as kugel (
22:53:05 Quit kugel (Nick collision from services.)
22:53:07TheSevenn1s: Actually I have no idea in what shape this stuff currently is
22:53:11 Join kugel [0] (
22:53:21 Quit kugel (Nick collision from services.)
22:53:27 Join kugel [0] (
22:53:39 Join kugel_ [0] (
22:53:50 Part Blue_Dude_
22:53:52 Quit kugel (Nick collision from services.)
22:53:55 Nick kugel_ is now known as kugel (
22:54:07 Quit kugel (Nick collision from services.)
22:54:13 Join kugel [0] (
22:54:25 Join kugel_ [0] (
22:54:36 Quit kugel (Nick collision from services.)
22:54:39 Nick kugel_ is now known as kugel (
23:04:00 Quit bertrik ("De groeten")
23:04:42 Quit evilnick ("Page closed")
23:08:20 Quit Omlet05 ("( :: NoNameScript 4.22 :: )")
23:10:05 Join domonoky1 [0] (
23:14:48 Quit Blue_Dude (Read error: 110 (Connection timed out))
23:18:39 Join rotator [0] (
23:20:44 Quit rotator (Client Quit)
23:21:10 Join rotator [0] (
23:21:45n1ssaratoga: question about r22871, why not just use memset ?
23:25:56 Quit evilnick_home (
23:25:56 Quit Thundercloud (
23:25:56 Quit Res1 (
23:25:56 Quit AndyI (
23:25:56 Quit xavieran (
23:25:56 Quit michaelcarr (
23:25:56 Quit antil33t (
23:25:56 Quit ej0rge (
23:25:56 Quit fish_ (
23:25:56 Quit bekabug (
23:25:56 Quit YPSY (
23:25:56 Quit timc (
23:25:57 Quit SUSaiyan (
23:25:57 Quit BlakeJohnson86 (
23:25:57 Quit redfox (
23:25:57 Quit jds2001 (
23:25:57 Quit rotator (
23:25:57 Quit FOAD (
23:25:57 Quit Horscht (
23:25:57 Quit Kohlrabi (
23:25:57 Quit faemir (
23:25:57 Quit Zarggg_ (
23:25:57 Quit shodanX (
23:25:57 Quit GeekShadow (
23:25:57 Quit mt (
23:25:57 Quit aidy (
23:25:57 Quit domonoky (
23:25:57 Quit howell_ (
23:25:57 Quit Unhelpful (
23:25:57 Quit pixelma (
23:25:57 Quit Zambezi (
23:25:57 Quit tchan (
23:25:57 Quit shadearg (
23:25:57 Quit feisar-_ (
23:25:57 Quit crwl (
23:25:57 Quit courtc (
23:25:57 Quit Overand (
23:25:57 Quit linuxguy3 (
23:25:57 Quit jvd (
23:25:57 Quit jfc (
23:25:57 Quit pjm0616 (
23:25:57 Quit Dhraakellian (
23:25:57 Quit kadoban (
23:25:57 Quit jordan` (
23:25:57 Quit tarbo (
23:25:59 Join jordan`` [0] (i=gromit@
23:25:59 Join linuxguy4 [0] (
23:25:59 Join The3_14ed|r [0] (
23:26:05NJoinjfc [0] (
23:26:13NJointarbo [0] (n=me@unaffiliated/tarbo)
23:26:36 Join jvd_ [0] (n=syscrash@
23:26:36NJoinrotator [0] (
23:26:36NJoinmichaelcarr [0] (
23:26:36NJoinThundercloud [0] (
23:26:36NJoinGeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
23:26:36NJoinHorscht [0] (n=Horscht2@xbmc/user/horscht)
23:26:36NJoinfaemir [0] (n=faemir@
23:26:36NJoinAndyI [0] (n=pasha_in@
23:26:36NJoindomonoky [0] (n=Domonoky@rockbox/developer/domonoky)
23:26:36NJoinmt [0] (n=MTee@rockbox/developer/mt)
23:26:36NJoinxavieran [0] (n=xavieran@
23:26:36NJoinantil33t [0] (n=Mudkips@
23:26:36NJoinevilnick_home [0] (
23:26:36NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
23:26:36NJoinZambezi [0] (i=Zulu@
23:26:36NJoinfeisar-_ [0] (
23:26:36NJoinaidy [0] (
23:26:36NJoinpixelma [0] (i=quassel@rockbox/staff/pixelma)
23:26:36NJoinFOAD [0] (
23:26:37NJoinRes1 [0] (
23:26:37NJoinZarggg_ [0] (
23:26:37NJoinshadearg [0] (
23:26:37NJoinKohlrabi [0] (
23:26:37NJoinUnhelpful [0] (n=quassel@rockbox/developer/Unhelpful)
23:26:37NJoinhowell_ [0] (
23:26:37NJoinYPSY [0] (
23:26:37NJoinfish_ [0] (
23:26:37NJoinej0rge [0] (
23:26:37NJoinjds2001 [0] (n=jds2001@fedora/jds2001)
23:26:37NJoinbekabug [0] (
23:26:37NJoinSUSaiyan [0] (
23:26:37NJoinredfox [0] (
23:26:37NJointimc [0] (n=aoeu@
23:26:37NJoinBlakeJohnson86 [0] (
23:26:37NJoinshodanX [0] (
23:26:37NJoinOverand [0] (
23:26:37NJoincrwl [0] (
23:26:37NJoincourtc [0] (n=court@unaffiliated/courtc)
23:26:40 Join kadoban [0] (n=mud@
23:26:57 Join SIGSEGV2 [0] (n=user@
23:26:57 Quit domonoky (Read error: 110 (Connection timed out))
23:28:26 Quit Llorean (Read error: 104 (Connection reset by peer))
23:29:04 Join Llorean [0] (
23:29:08 Quit Zarggg_ ()
23:29:35 Join Zarggg [0] (
23:30:51CtcpVersion from freenode-connect!freenode@freenode/bot/connect
23:30:52***Server message 477: 'logbot #rockbox :[freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff:'
23:34:59 Quit n1s ("Lämnar")
23:46:32 Quit Utchybann ("I like core dumps")
23:49:30 Join dfkt_ [0] (i=dfkt@unaffiliated/dfkt)
23:51:51 Join froggyman [0] (
23:53:35 Join nosa-j [0] (
23:55:41nosa-jhey if the ipod nano's seccond gen player can run rockbox will the 4th gen chronomatic nano be abel to do some things?
23:56:17nosa-jand if so is the work comeing along ok im so sick of the nano's firmware :D
23:58:26nosa-jfor one it never turn's completly off unless you restart the ipod and go into the configuration menu and hit hard sleep
23:58:37nosa-jso stupid XD

Previous day | Next day