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 2009-11-25

00:00:24kugelhm, the theme site is already over 8 month old...time passes quickly
00:00:41pixelmayou will need to hack the theme site either way
00:00:54pixelmayou means "someone" here
00:00:57minuscan you run gnome on rockbox?
00:01:24JdGordon|... the theme site knows which settings are reuquired.. so fixing that to enforce it makes more sense than fudging with the rockbox binary
00:01:24bluebrotherminus: no, only kde5 will get ported
00:01:45minusoh, cool
00:01:51pixelmaminus: you are aware that Rockbox is not linux?
00:02:02JdGordon|it.. NOT?!!!
00:02:05bluebrotherpixelma: Rockbox is not Linux?
00:02:07minusnot? what then? unix?
00:02:14AlexPIt is built from scratch
00:02:20bluebrotherWindows is Unix :P
00:02:26bluebrotherand Rockbox is Windows
00:02:43JdGordon|therefore rockbox is windows
00:02:43pixelmabluebrother: I thought it was DOS?
00:02:53minusbeen crawling through the docs though havent stumbled over the API yet
00:03:02bluebrotherpixelma: wasn't that changed when Cabbiev2 got the default theme?
00:03:17minuscabbiev2 is default theme?!
00:03:19pixelmafor Windows?
00:03:19amiconnMeh, can of worms :(
00:03:25AlexPminus: yes
00:03:25bluebrotheryou'll find the API documentation in the code ... as with every good OS project *g*
00:03:34minusi like rockboxed :3
00:03:54amiconnThis diacritics commit does more wrong than it does right
00:04:14JdGordon|and it was on FS for how long?
00:04:34Torneto be fair amiconn did point out the drawmode thing before when it was first discussed
00:04:37JdGordon|1 month apparently
00:05:00amiconnLike, what happens if one of the diacritics is wider than the base char? The diacritics will extend across the previous and next char
00:05:04bluebrotherand what does this time-around-in-FS tell? Nothing.
00:05:36JdGordon|it means that amiconn had plenty of time to review the patch...
00:05:45AlexPdepends how much it was pointed out
00:05:53amiconnThis will cause te display being broken, and is also probably not what the unicode standard says about combining
00:05:55minusso rockbox is not linux-based because the kernel would've been too large and other reasons aswell?
00:06:00*bluebrother notes that people *do* have a real life ...
00:06:10AlexPIf it was just there, you can't expect people to look at everything on there constantly
00:06:14amiconnJdGordon|: You expect all devs to check every new task in fs?
00:06:16bluebrotherplus other area of interests.
00:06:20gevaertsminus: why would that be? Can't a system be not linux without a special reason?
00:06:21*amiconn only checks fs when being pointed to
00:06:45amiconnMuch like the forums
00:06:45*JdGordon| was under the imppression that devs are expected to be on the fs email list...
00:07:10bluebrothera task on FS has gotten some review if there are a bunch of comments, not if the task was around for x weeks.
00:07:28JdGordon|and that task had plenty of comments
00:07:34amiconnI didn't even know there was such a patch
00:07:34bluebrotherJdGordon|: and you read every task?
00:07:38pixelmaI'm not
00:07:49JdGordon|the ones that look interesting
00:07:58*JdGordon| isnt the one complainig post commit
00:08:03bluebrotherwell, at least *I* don't have the *time* to check every taskk.
00:08:09amiconnI thought about combining marks myself quite some time ago. One reason why I didn't do it was that can of worms
00:08:18 Join rrrrex [0] (
00:08:22pixelmaI only get the open tasks mails through the dev ml every friday/saturday night
00:08:30 Join rrrrexxxx [0] (
00:08:30minusgevaerts, linux would've been a lazy way, and a reason for doing your own OS is wanting to write your own OS
00:08:37AlexPnot true
00:08:48AlexPlinux wouldn't have worked
00:08:48*amiconn never reads more than the number of open tasks in those mails
00:09:10gevaertsminus: look at ipodlinux, and see how far they got as an audio player
00:09:10AlexPlook at the resources of the first players Rockbox ran on, and of many of the ones now
00:09:20Torneamiconn: weren't you here on irc when the original thai-only version was discussed?
00:09:28Tornei'm pretty sure i remember you raising exactlyt hose objections then
00:09:33Tornesomeone did, anyway
00:09:49amiconnMaybe, I probably forgot
00:09:54minusAlexP, yes, but if it worked... i know the linux kernel is huge
00:10:00AlexPit doesn't
00:10:15AlexPhave a look at the players Rockbox runs on
00:11:23minusyes, ram
00:11:34minusand maybe some unspported archs
00:11:45AlexPlinux just isn't practical
00:12:02JdGordon|also.. using the RL excuse doesnt work when you are in IRC all day... if you have enough time to complain you have enough time to review a dozen email subject lines a day...
00:12:08amiconnminus: It's also the concept of linux that doesn't fit. At all
00:12:45minusthat would be?
00:13:03amiconnOne important point is memory management
00:13:17 Quit rrrrex ("CGI:IRC (Ping timeout)")
00:13:32pixelmaJdGordon|: you can't comparing reading a few lines of chat with reading code
00:13:32pixelmacompare too
00:13:55minusi dont know how memory management works on linux and on DAPs
00:14:06amiconnLinux does dynamic memory allocation. For one that means there needs to be always some free memory for programs to allocate from. But this is a waste: We want to buffer as much audio as possible, in order to keep hdd spinup time to a minimum
00:14:09JdGordon|no, but you can compare reading irc to reviing the *subject* line for the flyspray emails.. which is all I do
00:14:34JdGordon|when an interesting task coems up, then you can reply there with a comment saying that you'd like time to review before they commit
00:14:34 Quit rrrrexxxx ("CGI:IRC (EOF)")
00:14:38amiconnSo our buffer always needs to be full... something that causes huge problems with dynamic allocation
00:14:43bluebrotherJdGordon|: just because my client is running all day doesn't mean that I'm around all day...
00:14:47bluebrotherJdGordon|: just because my client is running all day doesn't mean that I'm around all day...or that I read all of the logs
00:15:04bluebrotherwhops, some enter press got lost.
00:15:14amiconnAnother, also memory management related problem is that dynamic allocation causes memory fragmentation on systems without an mmu
00:15:23JdGordon|I accept that.. but you both are in IRC pretty regularly
00:15:25amiconnAnd almost all rockbox targets have no mmu
00:15:55minussansa AMS models do :)
00:16:09amiconn(that also means that you can't run full linux but are restricted to uclinux)
00:16:27*bluebrother leaves for ... 20 hours or so for RL stuff
00:16:35amiconnminus: Read up on Rockbox' history, and have a look on the number of targets rockbox runs on
00:16:37minusbluebrother, sleeping? :D
00:17:06minusi think i will, this is definitely interesting
00:21:44*JdGordon| thinks bluebrother is missing the point
00:22:59 Join dude187 [0] (
00:27:24Incognito-AWAYum...out of curioisity
00:27:39Incognito-AWAYis the japanese language broken?
00:28:03JdGordon|I'm sure there are lynguists who would agree
00:28:22Incognito-AWAYoddly enough any of my songs that have japanese characters wont show up right...and changing the language to japanese has it show nothing in the menu
00:28:41Incognito-AWAYJDGordon| jokes aside i mean language support on rockbox
00:29:24TheSevenyou're probably just using a font that doesn't support japanese characters?
00:29:56*TheSeven isn't sure which fonts do support them though
00:29:58Incognito-AWAYwhich font on rockbox does support japanese characters? ( if any )
00:29:58Tornemost of the provided fonts have only regular western european chars
00:30:18amiconnUnifont definitely does
00:30:50amiconnrasher has a nice table showing character coverage of the various fonts
00:31:41froggymanIncognito-AWAY: you can also make your own font">
00:32:59Incognito-AWAYfroggyman, im tempted to
00:33:47amiconnLooks like Sazanami Mincho and the larger versions of 'Fixed' should support Japanese as well
00:34:13 Join Strife89 [0] (
00:35:12Incognito-AWAYyup yup
00:35:50Incognito-AWAYsize 12 Sazanami Mincho
00:35:55Incognito-AWAYthanks amiconn
00:35:58 Nick Ypsy is now known as YPSY (
00:41:45 Quit liar (Read error: 104 (Connection reset by peer))
00:45:58 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
00:50:50 Part froggyman
00:51:09 Quit JdGordon| ("Miranda IM! Smaller, Faster, Easier.")
00:53:52 Quit minus (Read error: 113 (No route to host))
00:58:10 Join froggyman [0] (
01:04:12 Join GeekShado_ [0] (
01:06:41 Part froggyman
01:11:33 Quit DerPapst ("Leaving.")
01:11:41 Quit GeekShadow (Read error: 148 (No route to host))
01:20:20***Saving seen data "./dancer.seen"
01:20:58 Quit efyx_ (Remote closed the connection)
01:22:48 Quit hebz0rl ("Ex-Chat")
01:26:03 Quit AaronM ("Emo Time In My Corner... //_-")
01:37:39 Part toffe82
01:39:04 Quit stripwax ("")
01:39:53 Quit dude187 (Read error: 113 (No route to host))
01:43:57 Quit kugel (Remote closed the connection)
01:47:44 Quit Thundercloud (Remote closed the connection)
01:53:32 Join froggyman [0] (
01:59:24 Part froggyman
02:00:10 Nick fxb is now known as fxb__ (
02:00:10 Nick fxb__ is now known as fxb (
02:00:10 Quit fxb (Success)
02:00:24 Join fxb [0] (
02:00:55 Nick fxb is now known as fxb__ (
02:11:46 Quit FlynDice (Remote closed the connection)
02:12:56 Join FlynDice [0] (
02:22:06 Quit Stephen_ ("Leaving")
02:28:52 Quit panni_ (Read error: 104 (Connection reset by peer))
02:35:20 Quit GeekShado_ (Read error: 104 (Connection reset by peer))
02:44:42 Join CaptainKewl [0] (
02:56:51 Join Lucky7hirteen [0] (n=freaking@
03:02:08 Quit JdGordon ("Leaving.")
03:02:28 Quit saratoga (Ping timeout: 180 seconds)
03:03:45 Join darkham [0] (
03:04:26 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
03:15:13 Quit n17ikh ()
03:15:14Incognito-AWAYanother out of curioisity question is....what happened to the link of the rockbox source on the main page?
03:15:21Incognito-AWAYwell atleast on the daily builds section
03:16:12 Quit MethoS- (Remote closed the connection)
03:16:21UnhelpfulIncognito-AWAY: it's preferable to use svn, really. you can update to a new revision that way without downloading the whole thing.
03:17:47Unhelpfulalso there is a task for AA fonts on rockbox, and i have plans for subpixel-AA. you'll need to generate entirely new fonts from truetype source fonts, so that they can be properly rendered for these new features. our usual source fonts are BDF bitmaps fonts, so they can't be made AA.
03:18:19 Part martian67
03:20:23***Saving seen data "./dancer.seen"
03:26:14 Join n17ikh [0] (n=n17ikh@
03:43:10 Part cendres
03:49:51 Quit cowgarden (Read error: 104 (Connection reset by peer))
03:53:09 Quit darkham ("Sto andando via")
04:03:34 Join FOAD_ [0] (
04:04:13 Join froggyman [0] (n=sopgenor@
04:06:43 Quit TheSeven (Nick collision from services.)
04:07:00 Join The_Seven [0] (n=theseven@rockbox/developer/TheSeven)
04:07:12 Nick The_Seven is now known as TheSeven (n=theseven@rockbox/developer/TheSeven)
04:17:56 Quit FOAD_ (
04:17:56 Quit JdGordon (
04:17:56 Quit FlynDice (
04:17:56 Quit grndslm (
04:17:56 Quit antil33t (
04:17:56 Quit goffa (
04:17:56 Quit kadoban_ (
04:17:56 Quit sbhsu (
04:17:56 Quit tha (
04:17:56 Quit togetic (
04:17:56 Quit AlexP (
04:17:56 Quit Incognito-AWAY (
04:17:56 Quit Dhraakellian (
04:17:56 Quit BlakeJohnson86 (
04:17:56 Quit Zambezi (
04:17:56 Quit xavieran (
04:17:56 Quit bzed (
04:17:56 Quit ps-auxw (
04:17:56 Quit fish_ (
04:17:56 Quit rjg (
04:17:56 Quit markun (
04:17:56 Quit hatseflats (
04:17:56 Quit Overand (
04:17:56 Quit knittl (
04:17:56 Quit pixelma (
04:17:56 Quit gevaerts (
04:17:56 Quit solexx_ (
04:17:56 Quit linuxguy3 (
04:17:56 Quit Res1 (
04:17:56 Quit tmzt (
04:17:56 Quit CaptainKewl (
04:17:56 Quit Horscht (
04:17:56 Quit FOAD (
04:17:56 Quit chaos (
04:17:56 Quit Torne (
04:17:56 Quit bluebrother (
04:17:56 Quit Rob2223 (
04:17:56 Quit tchan (
04:17:56 Quit Trista281 (
04:17:56 Quit scorche|sh (
04:17:56 Quit YPSY (
04:17:56 Quit jds2001 (
04:17:56 Quit tarbo (
04:17:56 Quit CIA-80 (
04:17:56 Quit froggyman (
04:17:56 Quit fxb__ (
04:17:56 Quit fyrestorm (
04:17:56 Quit amiconn (
04:17:56 Quit Sajber^ (
04:17:56 Quit phanboy4 (
04:17:56 Quit mikroflops (
04:17:56 Quit Utchybann (
04:17:56 Quit Tuplanolla (
04:17:56 Quit avacore (
04:17:56 Quit Lss (
04:17:56 Quit rphillips (
04:17:56 Quit rvvs89 (
04:17:58 Quit killan (
04:17:58 Quit alexbobp (
04:17:58 Quit advcomp2019 (
04:17:58 Quit linuxstb (
04:17:58 Quit scorche (
04:17:58 Quit jfc^3 (
04:17:58 Quit bughunter2 (
04:17:58 Quit niekie_ (
04:17:58 Quit preglow (
04:17:58 Quit maraz (
04:17:58 Quit crashd_ (
04:17:58 Quit crwl (
04:17:58 Quit Unhelpful (
04:17:58 Quit TaZzZ (
04:17:58 Quit krazykit (
04:17:58 Quit blithe (
04:17:58 Quit B4gder (
04:17:58 Quit lyngaas (
04:17:58 Quit jvd (
04:17:58 Quit ehntoo (
04:17:58 Quit z35 (
04:17:58 Quit TheSeven (
04:17:58 Quit Lucky7hirteen (
04:17:58 Quit Topy44 (
04:17:58 Quit topik (
04:17:58 Quit gtkspert_ (
04:17:58 Quit kkurbjun (
04:17:58 Quit Galois (
04:17:58 Quit mc2739 (
04:17:58 Quit elcan (
04:17:58 Quit zu_ (
04:17:58 Quit Kohlrabi (
04:17:58 Quit Zarggg (
04:17:58 Quit hd (
04:17:58 Quit SIGSEGV (
04:19:02NJoinTheSeven [0] (n=theseven@rockbox/developer/TheSeven)
04:19:02NJoinfroggyman [0] (n=sopgenor@
04:19:02NJoinLucky7hirteen [0] (n=freaking@
04:19:02NJoinCaptainKewl [0] (
04:19:02NJoinfxb__ [0] (
04:19:02NJoinfyrestorm [0] (
04:19:02NJoinHorscht [0] (n=Horscht2@xbmc/user/horscht)
04:19:02NJoinamiconn [0] (i=quassel@rockbox/developer/amiconn)
04:19:02NJoinpixelma [0] (i=quassel@rockbox/staff/pixelma)
04:19:02NJoinSajber^ [0] (
04:19:02NJoinphanboy4 [0] (
04:19:02NJoinsolexx_ [0] (
04:19:02NJoinFOAD [0] (
04:19:02NJoinTopy44 [0] (
04:19:02NJoinRob2223 [0] (
04:19:02NJoinbluebrother [0] (n=dom@rockbox/developer/bluebrother)
04:19:02NJoinavacore [0] (i=nobody@
04:19:02NJoinTaZzZ [0] (
04:19:02NJoinhd [0] (n=jd@Wikipedia/HellDragon)
04:19:02NJoinalexbobp [0] (n=alex@
04:19:02NJoinLss [0] (
04:19:02NJoinlinuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
04:19:02NJointarbo [0] (n=me@unaffiliated/tarbo)
04:19:02NJoinmikroflops [0] (
04:19:02NJoinkillan [0] (
04:19:02NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
04:19:02NJoinSIGSEGV [0] (n=user@
04:19:02NJoinUtchybann [0] (
04:19:02NJoingtkspert_ [0] (
04:19:02NJoinZarggg [0] (
04:19:02NJoinniekie_ [0] (
04:19:02NJoinlinuxguy3 [0] (
04:19:02NJoinscorche [50] (n=scorche@rockbox/administrator/scorche)
04:19:02NJoinTorne [0] (i=torne@rockbox/developer/Torne)
04:19:02NJoinjfc^3 [0] (
04:19:02NJoinmc2739 [0] (n=mc2739@rockbox/developer/mc2739)
04:19:02NJointmzt [0] (
04:19:02NJoinRes1 [0] (
04:19:02NJoinbughunter2 [0] (n=bughunte@unaffiliated/bughunter2)
04:19:02NJoinYPSY [0] (
04:19:02NJoinblithe [0] (
04:19:02NJoinehntoo [0] (
04:19:02NJoinrvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
04:19:02NJoinGalois [0] (
04:19:02NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
04:19:02NJoinz35 [0] (
04:19:02NJoinkrazykit [0] (
04:19:02NJoinjds2001 [0] (n=jds2001@fedora/jds2001)
04:19:02NJoinkkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
04:19:02NJoinzu_ [0] (
04:19:02NJoincrashd_ [0] (
04:19:02NJoinmaraz [0] (
04:19:02NJointopik [0] (
04:19:02NJoinchaos [0] (n=chaos@gentoo/user/ch4os)
04:19:02NJoinpreglow [0] (
04:19:02NJoinlyngaas [0] (
04:19:02NJoinTrista281 [0] (
04:19:02NJoinCIA-80 [0] (n=CIA@
04:19:02NJoinrphillips [0] (
04:19:02NJoinelcan [0] (
04:19:02NJoincrwl [0] (
04:19:02NJoinB4gder [241] (n=daniel@rockbox/developer/bagder)
04:19:02NJoinKohlrabi [0] (
04:19:02NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
04:19:02NJoinUnhelpful [0] (n=quassel@rockbox/developer/Unhelpful)
04:19:02NJoinjvd [0] (n=syscrash@poipu/developer/syscrash)
04:19:02NJoinadvcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
04:20:22 Quit FOAD (Read error: 110 (Connection timed out))
04:24:32CIA-80New commit by FlynDice (r23746): AMS Sansa: For consistency, replace references to SECTOR_SIZE with SD_BLOCK_SIZE which is defined in sd.h as 512 bytes. ...
04:26:48NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
04:27:40 Quit Tomis (Connection timed out)
04:36:12 Quit gevaerts (Nick collision from services.)
04:36:24 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
04:37:17NJoinJdGordon [0] (n=jonno@rockbox/developer/JdGordon)
04:37:29 Part froggyman
04:42:00NJoingoffa [0] (n=goffa@
04:48:57 Quit JdGordon ("Leaving.")
04:51:14 Join JdGordon [0] (
04:54:59 Quit CaptainKewl ("( :: NoNameScript 4.22 :: )")
05:02:57n17ikhis there a bug in the time/date setting screen, or is the RTC in my e260 broken? every time I go to set the time it advances a day past what I had set when I press the set button
05:14:01 Join JdGordon1924 [0] (
05:16:02 Quit Horscht ("Verlassend")
05:18:59 Quit JdGordon (Read error: 110 (Connection timed out))
05:20:27***Saving seen data "./dancer.seen"
05:20:41 Join StealthyXIIGer [0] (n=stealthy@
05:22:14 Join ShapeShifter499 [0] (
05:24:39 Quit Lss (Read error: 104 (Connection reset by peer))
05:32:22 Quit Strife89 (Read error: 104 (Connection reset by peer))
05:32:53 Join Strife89 [0] (
05:34:34 Join FOAD [0] (
05:45:31 Quit Zarggg (Read error: 104 (Connection reset by peer))
05:55:43 Join Zarggg [0] (
06:00:39 Quit Strife89 ("Bed.")
06:34:42 Quit ShapeShifter499 (Read error: 104 (Connection reset by peer))
06:39:34 Quit JdGordon1924 (Read error: 110 (Connection timed out))
06:46:38 Join JdGordon [0] (
06:51:41 Quit shaggy-h (Read error: 110 (Connection timed out))
06:58:59 Quit bluebrother (Nick collision from services.)
06:59:01 Join bluebroth3r [0] (n=dom@rockbox/developer/bluebrother)
07:12:03 Join matsl [0] (
07:18:37 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
07:20:29***Saving seen data "./dancer.seen"
07:27:39 Join Tomis [0] (n=Tomis@
07:27:39 Join FlynDice [0] (
07:27:39 Join antil33t [0] (
07:27:39 Join xavieran [0] (
07:27:39 Join tha [0] (
07:27:39 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
07:27:39 Join kadoban_ [0] (
07:27:39 Join sbhsu [0] (
07:27:39 Join BlakeJohnson86 [0] (
07:27:39 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
07:27:39 Join Zambezi [0] (i=Zulu@
07:27:39 Join Dhraakellian [0] (
07:28:12 Join bzed [0] (
07:28:12 Join ps-auxw [0] (
07:28:12 Join hatseflats [0] (n=hatsefla@
07:28:12 Join rjg [0] (
07:28:12 Join knittl [0] (n=knittl@unaffiliated/knittl)
07:28:12 Join fish_ [0] (
07:28:12 Join Overand [0] (
07:28:12 Join markun [50] (n=markun@rockbox/developer/markun)
07:28:48 Join slam_ [0] (n=grndslm@
07:29:14 Nick slam_ is now known as grndslm (n=grndslm@
07:29:34 Quit StealthyXIIGer (Read error: 110 (Connection timed out))
07:52:44 Quit hd (Read error: 104 (Connection reset by peer))
07:54:48 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:58:34 Join HellDragon [0] (
08:25:42 Join Grahack [0] (
08:45:37 Quit Grahack ("Leaving.")
08:49:08 Join Rob2222 [0] (
08:55:36 Join Grahack [0] (
08:57:34 Join maruk [0] (n=papier@
09:00:05 Join petur [50] (n=petur@rockbox/developer/petur)
09:07:09 Quit Rob2223 (Read error: 110 (Connection timed out))
09:08:28n17ikhany clues on why I have to reboot twice to get the database to show changed files?
09:13:29 Join flydutch [0] (
09:14:01 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
09:19:11 Join dude187 [0] (
09:20:32***Saving seen data "./dancer.seen"
09:23:24 Join Thundercloud [0] (
09:23:58 Join Bagder [0] (
09:31:55 Join einhirn [0] (
09:36:20 Quit linuxstb (Read error: 110 (Connection timed out))
09:37:21 Join DerPapst [0] (n=DerPapst@
09:46:19 Quit Thundercloud (Remote closed the connection)
09:46:39 Quit JdGordon (Read error: 104 (Connection reset by peer))
09:53:17 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
09:56:27 Quit antil33t (Read error: 104 (Connection reset by peer))
10:01:16 Join ShapeShifter499 [0] (
10:05:21 Join funman [0] (n=fun@rockbox/developer/funman)
10:16:36 Quit Llorean (Read error: 131 (Connection reset by peer))
10:26:39 Quit DerPapst ("Leaving.")
10:31:33Zagorphilosophy time. what is a port? are ipodvideo and ipodvideo64mb different ports? they run the same code
10:31:45 Join DerPapst [0] (
10:32:03Bagderwell, not exactly the same...
10:32:32Bagderbut yes, it's certainly not that easy to answer
10:33:02Tomisa port is when you have to change the software in order to make it run properly
10:33:12Tomisotherwise, it's just a recompile for a different target
10:33:12funmanfrom a porter point of view, i would say clipv1, m200v4, c200v2, e200v2 and fuze are the same port considering what they have in common
10:34:13funmaneven if there are differences between each (and each have specific code), they are minimal compared to what they have in common: there is more differrences between a fuze and an ipod than between a fuze and a clip
10:37:04Bagderfunman: so how big differences can there be and you still call it the same port?
10:37:30funmanleft as appreciation to the philosopher
10:44:53BagderI would rather say that if you need to figure out something specific for a device and use a different code for it, it is a new port. Thus I'd call those sansa devices different ports.
10:49:24 Quit DerPapst (Read error: 104 (Connection reset by peer))
10:49:42ZagorI guess what I'm doing is rather "builds" than "ports"
10:50:12Bagderyeah, that's an easier definition I think
10:52:06 Quit ShapeShifter499 ("ChatZilla 0.9.85 [Firefox 3.5.5/20091109125225]")
10:52:18 Quit phanboy4 (Read error: 113 (No route to host))
10:54:15Zagorwhat is the onda 747p? I find vere few mentions of it online
10:55:04 Join DerPapst [0] (n=DerPapst@
10:56:48BHSPitLappyI would say a port is a common location where vessels make harbor, for loading/unloading passengers and goods
10:57:23BHSPitLappyalternately a software abstraction for network-layer internet protocol
10:58:26dionoeahowdy. funman: you don't have a (working) clip anymore ?
10:59:14funmanyes i do
10:59:59BagderZagor: I think it is "747+"
11:00:21funmandionoea: i won't test your patch because I can't reproduce easily crashes on my clip, it looks much simpler on c200v2
11:00:35dionoeaah ok. betrik has a c200v2 ?
11:00:38funmanby the way you can simulate a sansa clip in your own build by lowering the size of audio buffer
11:01:03*BHSPitLappy doesn't get no respect
11:01:05*dionoea might have forgoten a letter in that nick
11:01:17BHSPitLappynone at all...
11:01:41funmandionoea: just add a static char bla[5<<20]; in rockbox (you might want to tune the value)
11:01:48funmanand yes bertrik has a c200v2
11:03:57dionoeafunman: ok. I'll try that tonight
11:04:21dionoeadoes the c200v2 have an even smaller buffer than the clip?
11:04:34funmanyes, mostly due to having a color lcd
11:05:07Bagder>2 color lcd ;-)
11:06:12funmanwell technically the Clip has no LCD at all ;)
11:08:07 Quit dude187 (Read error: 113 (No route to host))
11:11:06 Join DerPapst1 [0] (n=DerPapst@
11:12:17 Quit TheSeven (Read error: 113 (No route to host))
11:20:35***Saving seen data "./dancer.seen"
11:25:53Zagoramiconn, pixelma: are you using the 8MB archos recorder builds?
11:26:02Bagdersorry, my fault!
11:26:16Bagderoh, wrong channel
11:26:22*Bagder falls over
11:27:00 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
11:31:22 Quit DerPapst (Read error: 110 (Connection timed out))
11:31:49funmanwhat was decided at the gsoc mentor summit about integrating back (some) rockbox modifications of ffmpeg in their svn (especially wma codec)
11:38:58funmani'll ask again when saratoga is here
11:39:14Bagderyeah, he's the man with clues on those parts
11:39:41 Quit funman ("free(random());")
11:40:30 Join DerPapst [0] (
11:49:25linuxstbZagor: Don't we use the term "target" already? Meaning one of the options in tools/configure. So currently "ipodvideo" is one "target", but with two "builds". I've been wondering if they should be two distinct targets, but the problems are that there is only one bootloader, and it's not easy for a user to know what RAM size they have.
11:49:42 Quit DerPapst1 (Read error: 104 (Connection reset by peer))
11:50:51Torneis anything actually different other than the audio buffer size for those, btw?
11:51:43Bagderthe codecs and plugns are placed after the audio buffer...
11:54:21Torneis there a reason for that? :)
11:54:27linuxstbSo their location is fixed.
11:55:04*Torne doesn't get it
11:55:14*gevaerts suffers from deja vu :)
11:55:21Tornesurely their location would still be fixed if it was before the audio buffer
11:55:31linuxstbNo, because Rockbox builds vary in size.
11:55:50linuxstbI mean fixed across different Rockbox builds.
11:55:53TorneOhh, of course
11:55:56Bagderand plugins have a fixed size
11:56:26Tornei should really remember this having explained to people before the conditions under which plugins are backward/forward compatible
11:56:40linuxstbThe obvious solution is to move the codecs and plugins _before_ the main Rockbox binary, but that needs someone to do it...
11:56:53gevaertsor making them position independent
11:57:14Tornethey probably are mostly position independant already, on ARM
11:57:45linuxstbBut yes, this is one of those things we've been talking about for years, but no-one has done...
11:58:03Tornewas someone working on having the init code discarded, also?
11:58:35Bagderkugel has a patch for that he said
11:59:00Tornei am tempted to write the code to detect ram size anyway, is all
12:05:21 Join kugel [0] (i=kugel@rockbox/developer/kugel)
12:05:48kugeltogetic: FS #10756
12:05:54kugelTorne: ^^
12:09:15kugelre: fixed plugin/codec buffer location
12:09:23 Join DerPapst1 [0] (n=DerPapst@
12:09:23 Quit HellDragon (Read error: 54 (Connection reset by peer))
12:09:25 Join hd [0] (n=jd@
12:09:41kugelI was wondering today why we have this kind of api which rather restricts things. What if we had a syscall mechanism?
12:09:49 Quit DerPapst1 (Client Quit)
12:11:42Tornekugel: i'll have a look/play with that patch later (it interests me) :)
12:12:15kugelit works fine for e200v1 as far as I tested
12:12:23Tornei can test on several other devices
12:12:30Tornei'm acquiring an ever increasing set ;)
12:13:19Tornea syscall mechanism is one option.. not sure about other arches but the cost on ARM can be minimal
12:13:21kugelI've got a list of other (semi-)targets too, but I was busy lately so I didn't work on it further
12:13:33Torneit could be less if ARM was slightly smarter, alas
12:13:45gevaertskugel: the api doesn't restrict things as far as I can see, since it's just a big function pointer table which could be put anywhere. From what I understand there are two things that "fix" the location : entry point (plugin main) locaction (easy to fix I guess), and position dependent code
12:14:19Tornegevaerts: since everything is required ot be compiled per-target anywa, though, the advantage of PIC seems pretty tiny
12:14:35Tornethe memory rearrangement mentioned above seems a nicer solution
12:15:22Tornei guess it depends what things you want to free yourself up to vary; i was only considering total ram size since it seems an unfortunate thing to have to produce a seperate build for
12:15:32gevaertsTorne: what I mainly like about PIC is that it would allow for more than one plugin at the same time (there's more to that than just PIC of course, but it's the main one I think)
12:15:38kugelpic is rather easy to achieve on arm, isn't it (especially with eabi)?
12:16:13Tornekugel: yes. basically everything other than references to .data and function pointers are PIC anyway
12:16:37Tornegevaerts: Hmm
12:16:48gevaertsMore plugins allow things like USB driver plugins (keep MSC in core, but full HID? MTP? Audio?), WPS visualisation plugins,...
12:17:08gevaertsand of course the most important one : battery_bench can keep running while you play doom :)
12:17:17*Torne ponders how difficult it is to get ARM to believe that it can access .data pc-relatively
12:17:47kugelldr r0, pc+X ?
12:18:03Tornekugel: It's trivial in *assembly*, yes
12:18:16Tornebut the compiler is generally going to assume that inter-section references are absolute
12:18:27kugeland gcc won't generate that with -fPIC?
12:18:42Tornegcc will *normally* use a base register for data
12:18:53Tornebecause it assumes the linker may want to put .data too far away
12:19:00Torner9 gets used as sb, static base
12:19:06Torneand you lose a register.
12:19:21gevaertsWe need to check code size for PIC of course, especially on sh
12:20:16kugelTorne: that seems stupid :/
12:20:35Tornekugel: Why? On the vast majority of machines where C code is used, the .data section is not going to be close neough is not mapped immediately after .text on regular OSes
12:20:57Tornethere may be some way to convince it to do so
12:21:05Tornebut if so i don't know what it is offhand.
12:22:01kugelwell, on ARM it seems rather unlikely that .data is too far away (not accunting stuff like .idata)
12:22:15Torneon rockbox, sure.
12:22:41Torne-mno-word-relocations might be relevant, but is described poorly :)
12:23:38linuxstbBagder: Do you have any download stats on the 8MB Archos build(s) ?
12:28:33 Quit mc2739 (Read error: 110 (Connection timed out))
12:29:54Torneoh, wait. i can't count
12:30:05Torneload immediate offsets are only 12 bits plus sign
12:30:12Tornerange of +/-4095 bytes, useless.
12:31:07Torneso yes, you'd have to accept the general loss of a register and some ram for a GOT
12:32:25Tornewell maybe not a GOT, depends how it feels like doing it
12:34:55 Quit BHSPitLappy (Read error: 110 (Connection timed out))
12:35:11Torneglobal offset table
12:35:35 Quit DerPapst (Read error: 110 (Connection timed out))
12:37:50Tornebah, except it isn't likely to need a got either if it's been set up right, at least not on non-xip
12:38:08*Torne tries, instead. :)
12:39:45 Join minus` [0] (
12:47:38 Part Bagder
12:50:37 Quit Grahack ("Leaving.")
12:53:40 Join Bagder [0] (
12:54:08 Join Omlet [0] (
13:00:51Torneit does generate a GOT, bugger.
13:00:57Tornewhich our link scripts then choke on ;)
13:01:41Tornewow, this generated code *offends my eyes*
13:13:02 Quit einhirn (Read error: 60 (Operation timed out))
13:20:36***Saving seen data "./dancer.seen"
13:21:33 Join adiroiban [0] (
13:22:04kugelTorne: haha
13:22:11Tornekugel: it really does.
13:22:46kugelyes, I do believe that ;)
13:24:01Bagderalways wear protective goggles!
13:24:59 Part adiroiban
13:26:19 Join mc2739 [0] (n=mc2739@rockbox/developer/mc2739)
13:27:04Tornethe problem is i can *imagine* vastly more efficient ways ofr this code to work
13:27:17Tornebut it's hard to see how you could possibly convince gcc to comply :)
13:30:22Tornegevaerts: general result: what gcc generates on ARM when you tell it -fpic is vastly different (and spectacularly worse) than what comes to mind when you think of position independant code in the sense we require :)
13:30:53topikcan i build mkamsboot (win32) from a normal rockbox svn checkout?
13:31:28Tornehowever the optimal case for what we want is *probably* completely unachievable even by a reasonably sane level of patching gcc. i may have to ask someone.
13:31:47*Torne also needs to try eabi.
13:32:16 Join funman [0] (n=fun@rockbox/developer/funman)
13:32:20Bagdertopik: there is no different checkouts, only "the normal"
13:32:43Bagderso yes
13:33:08funmandionoea: i started using your last patch on my clip, will report if I see a crash
13:34:52topikcan't find the right option to build it, Bagder :(
13:34:58 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:35:44funmantopik: why do you want to build it yourself?
13:35:46kugeltopik: I don't think anyone tried building on windows yet, but have you tried make and mingw?
13:37:40topiki'm on linux, but i want to build the .exe
13:38:07topikfunman: because i'm under the impression the svn version works with the latest fuze firmware
13:38:09kugelyou need a cross compiler then
13:38:25BagderI bet our makefile is not written for that
13:38:35topikcompiling is set up fine. i can compile rockbox, bootloaders etc. just can't find the option to build the firmware patcher
13:38:58Bagderright, it's a bit hidden
13:39:13Bagdercd rbutil/mkamsboot && less README
13:39:55funman% make CC=i586-mingw32msvc-gcc mkamsboot.exe
13:40:12Bagderoh nice
13:41:12funmannext rbutil release will support this fuze firmware
13:42:20topiki'm on linux though, so i don't think CC=i586-mingw32msvc-gcc will fly
13:43:48funmanit's the name of the win32 cross compiler packaged by debian, but I gave you a link to a mkamsboot build which supports last fuze firmware, didn't you t
13:43:51funmanread it* ?
13:43:56topikyea i did
13:44:00topikand i appreciate that
13:45:30topiki was just curious about the process of compiling my own, but i will use saratoga's instead.
13:45:36topikthanks for the help sirs.
13:54:04 Join pamaury [0] (n=pamaury@
14:00:40kugelfunman: maybe you want to post a c200v2 build to the test forum? there seem to be some c200v2 user around at the moment
14:02:42funmani rely on our high quality c200v2 tester :)
14:06:11*Torne builds with eabi and noneabi with -fpic -msingle-pic-base and checks to see if one or the other makes him want to vomit and die *less*
14:09:40Unhelpful4.4.2 eabi is still as much as about 4% slower for some codecs
14:10:06funmansomeone mentioned gcc from codesourcery was more performant on ARM
14:12:44Tornefunman: codesourcery's version has Various Hilarity in it, yes
14:13:00Tornedunno how much difference it will make though
14:13:09Tornei should try RVCT sometime just for kicks
14:13:19Torne(and as a standard to compare to)
14:16:58funmangood luck for tweaking the makefiles to use rvct (didn't succeed)
14:18:02Unhelpful"CodeSourcery customers occasionally request that changes that they have sponsored not be contributed, often because the changes relate to CPUs that are not yet publicly available." <- surely that can't be compliant with the license if those changes are in a toolchain that codesourcery distributes?
14:18:39Bagdergcc is GPL isn't it?
14:18:52Unhelpfulit is.
14:19:06Bagderbut right, they don't distribute those binaries to anyone
14:19:20Torneindeed; they just don't give people the binaries who aren't special and magic
14:19:30Torneand thus the requirement to give those people the source doesn't actually result in public disclosure.
14:20:40Torneand yes, the eabi/noneabi versions of this both make me want to die. :)
14:22:03*Unhelpful is trying to figure out where you get the source for the not-for-special-people codesourcery toolchain
14:22:40Unhelpfulas we surely want to apply the same patches as for stock gcc, to get rid of exception handling and get the multilib binaries we want
14:24:11Tornethat's an arm-none-eabi toolchain
14:24:15Tornewhich is *probably* the one you want
14:24:27Tornebut may not behave the same as the one you compiled before :)
14:28:16Unhelpfulah, i think the link i thought would be to a package might've been to this page :)
14:28:35Torneyeah they aren't very helpful
14:30:46Unhelpfuli rather doubt there'd be any interest in switching to a gcc from some other party though...
14:33:21kugelTorne: does that create the .elf files?
14:33:39Tornekugel: yes, the "none" is not the binary format but OS
14:33:55 Quit minus` (Read error: 113 (No route to host))
14:33:57Tornethe toolchain Unhelpful's script mkaes is arm-elf-none-eabi really
14:34:07kugeland what is arm-elf[-eabi] then?
14:35:02Torneit's also arm-elf-none-eabi
14:35:21Tornethe various bits of the target spec are mostly optional and not delineated very consistently
14:35:58Torneit's not trivial to tell which targets are actually the same, but i'm *fairly* sure codesourcery's is the same target as unhelpful's :)
14:36:23Tornefor added fun the toolchain I use for pycorn is just arm-eabi, and is probably *also* the same ;)
14:43:18 Join webguest15 [0] (
14:44:10 Join froggyman| [0] (
14:44:13 Quit webguest15 (Client Quit)
14:45:13 Quit Topy44 (Read error: 104 (Connection reset by peer))
14:45:26 Join Topy44 [0] (
14:50:30 Quit FOAD (Read error: 110 (Connection timed out))
14:51:19 Join Llorean [0] (
15:00:40 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
15:00:59 Join FOAD [0] (n=dok@
15:03:29 Quit funman ("free(random());")
15:08:51Tornegevaerts: have you actually tried pic at all? *g* -fpic produces truly offensive code on ARM.
15:11:14kugelwhat -O do you use ?
15:13:05 Part Bagder
15:13:10gevaertsTorne: I never look at the generated code :)
15:13:19Tornekugel: doesn't matter
15:13:25Tornethe way the GOT works means it can never be better
15:13:51Tornewe would need -fembedded-pic or -mwithout-got or various other dodgy patches that uClinux used to maintain on 2.95 and have seemingly never been ported to 3.x let alone 4.x
15:13:56Tornewell i say need, i mean want
15:17:05Tornealso the whole GOT approach is probably useless to us anyway
15:17:19Tornesince it depends on a dynamic loader to do the relocs in the GOT
15:20:37***Saving seen data "./dancer.seen"
15:27:28 Part LinusN
15:27:47 Join watto [0] (n=watto@
15:28:53 Join Kaspar [0] (n=kaspar@
15:29:19 Nick fxb__ is now known as fxb (
15:32:37 Join T44 [0] (
15:34:40 Quit kugel (Read error: 110 (Connection timed out))
15:37:33 Quit Zagor ("Don't panic")
15:43:02 Join kugel [0] (i=kugel@rockbox/developer/kugel)
15:51:28 Quit Topy44 (Read error: 110 (Connection timed out))
15:51:59KasparThe sansa clip+ is currently available very cheaply here in Germany. It seems quite an interesting target, so if any developer is interested I would be willing to sponsor one (and helping out with my rather minor C skills). Is there anybody already working on it?
15:52:50 Quit froggyman| ("CGI:IRC (EOF)")
15:53:43Tornethe clipv2 is being worked on, the clip+ is *probably* almost identical hardware other than the uSD slot so it may be that much of the same code works
15:54:06Torne is the current clipv2 port
15:58:53 Quit parafin (Read error: 60 (Operation timed out))
16:02:42KasparI know, but is anyone actually working on a clip+ or willing to?
16:03:08Kaspar(and yes, I scanned the wiki&forums&IRC before...)
16:03:45 Join parafin [0] (
16:03:53 Join graey [0] (
16:04:09 Nick graey is now known as geertvdijk (
16:04:23geertvdijkhi all :)
16:04:24 Join LambdaCalculus37 [0] (n=LambdaCa@rockbox/staff/LambdaCalculus37)
16:04:42geertvdijkthis is the right channel for rockbox-related non-dev questions right?
16:05:38Lloreangeertvdijk: It's the right channel for any questions about Rockbox.
16:05:43geertvdijkgreat, thanks
16:06:04geertvdijkI have searched around a bit about my question already, and I found an answer, but then I lost it and I have no clue where I found it before
16:06:24geertvdijkI'm using the current build of rockbox and Windows wants to install a driver
16:06:46geertvdijknow I read somewhere on the site that you should push some button(s) and/or put it on hold, but I cant exactly remember and I cant find it either
16:06:56geertvdijkcan someone help me with that?
16:07:17LambdaCalculus37Which player?
16:07:27geertvdijksansa e250
16:07:32geertvdijknon-r, non-AMS
16:07:41kugelKaspar: I think many people would like to work on it, but usually lack of time/skills/real strong interest make lead to "nothing happens"
16:07:59LambdaCalculus37geertvdijk: And what are you looking to do? USB mode? Charge only?
16:08:00 Join DerPapst [0] (
16:08:08LambdaCalculus37Details, please.
16:08:09kugelYou're welcome to donate hardware, but don't count on anybody to work on it
16:08:10geertvdijkconnect as mass-storage
16:08:17LambdaCalculus37Just plug it in.
16:08:48geertvdijkI plugged it in already, windows is asking for drivers, and it doesnt do this with the stable build (3.4)
16:08:50LambdaCalculus37Or are you talking about it appearing to Windows as a USB HID device?
16:08:53geertvdijkHID mode is also working fine
16:09:03geertvdijkno, its appearing as both, but Windows wants drivers for the mass-storage part
16:09:25TorneKaspar: I don't think anyone is working on clip+, no, but since it's probably identical to clipv2 there is no particular need at the moment
16:09:42ThomasAHTorne: is someone working on clipv2 then?
16:09:54Torneyes, funman is, i linked to his FS# above.
16:10:58gevaertsgeertvdijk: which revision are you using?
16:11:08geertvdijkjust downloaded the latest
16:11:08 Join gb_master [0] (i=82c03212@gateway/web/freenode/x-ymeohrqmoevxjuxv)
16:11:08ThomasAHTorne: ah, good. I already saw that page but did not understand it :)
16:12:36gevaertswhat happens if you disable HID?
16:12:42geertvdijkhow do I do that?
16:13:06gevaertsin settings somewhere
16:13:14geertvdijkI'll unplug and find out, just a sec
16:13:24geertvdijkon a sidenote: I really love what you guys are doing at, any indication when it will become the main site? it's gorgeous and thats what rockbox deserves :)
16:14:52geertvdijkHID disabled: mass storage works
16:15:12gevaertswhich windows version is this?
16:15:23geertvdijkXP sp3 with all the available updates
16:16:09geertvdijkI also had it on my Vista Ultimate-machine though
16:16:21geertvdijkand my mom runs seven at her place, same issue.
16:16:30geertvdijkthat's windows 7 final ofcourse
16:17:51gevaertscan you submit a bug report? The person who works on HID (tomers) isn't here at the moment
16:21:18geertvdijkhow should I describe this?
16:21:35geertvdijke200 won't connect mass storage with HID enabled?
16:22:19kugelthat works in the current build, doesn't it?
16:23:38geertvdijkI'm running the current build, and for me it doesn't
16:24:56geertvdijkor werent you talking to me ^^
16:25:48 Join toffe82 [0] (n=chatzill@
16:26:16topikthat looks amazing
16:26:29mc2739kugel, geertvdijk: it does work in the current build, but on my e270r, it mysteriously started failed with hid enabled.
16:26:59mc2739my e280 still works with hid enabled
16:27:20geertvdijkso on your e270r it doesn't work with HID? Or it just didnt work one and after that it worked fine?
16:27:57mc2739geertvdijk: it did work with hid enabled and now it does not.
16:29:16geertvdijkso it didnt fail at all until I mentioned it? magic! :p
16:30:51mc2739geertvdijk: no, I started having the problem several days ago.
16:31:00geertvdijkoh ok
16:32:45mc2739I think the failure started after I connected the e270r using the OF in MTP mode, but I'm not positive that that is what caused the problem.
16:32:56geertvdijkoh, thats something worth checking out though
16:33:15geertvdijki'll check to see what mine is set up to use
16:33:35pixelmaI have no problems with my c200 so far on XP SP3 but after the last fix by gevaerts Windows wanted to install new drivers again, letting this go worked for me
16:34:12gevaertsdisabling HID will give you a different serial number
16:34:45geertvdijkmy sansa is set to MSC in OF and has always been I think
16:34:46gevaertsand Windows is buggy. It uses registry information even if that disagrees with the USB descriptors
16:34:59 Quit pamaury (Read error: 148 (No route to host))
16:35:02mc2739geertvdijk: even after switching back to MSC mode, I have problems. I planned to post a task on Flyspray, but have not been able to gather enough information yet,
16:35:12geertvdijkyeah, but it always was in MSC
16:35:28pixelmaI believe there was a report in the forums too
16:35:51gevaertsmc2739: if you do, please include all USB serial numbers you can find :) (rockbox with HID, rockbox without HID, OF MSC, OF MTP)
16:35:53geertvdijkI booted into it just to check and it already was in MSC, and I never use the OF, not for MTP either
16:36:03gevaertsmc2739: also the USB VID and PID
16:36:04mc2739As I said above, I am not sure that MTP mode caused the problem, it could have been just coincidental.
16:36:21geertvdijkyeah okay
16:36:34geertvdijkif I'm not mistaking that is probably not the case
16:36:53mc2739gevaerts: I will
16:37:01geertvdijkI'll check the serial numbers and VID/PIDs as well and add them to the bugreport
16:39:33DEBUGEOF from server (Connection reset by peer) (snapshot: netstuff.c line 545)
16:39:33***Saving seen data "./dancer.seen"
16:39:35***Started Dancer V4.16
16:39:35***Connected to on port 6667
16:39:35***Logfile for #rockbox started
16:39:45***Server message 501: 'logbot :Unknown MODE flag'
16:39:45Mode"logbot :+i" by logbot
16:39:45 Join logbot [0] (
16:39:45 Join toffe82 [0] (n=chatzill@
16:39:45 Join DerPapst [0] (
16:39:45 Join LambdaCalculus37 [0] (n=LambdaCa@rockbox/staff/LambdaCalculus37)
16:39:45 Join geertvdijk [0] (
16:39:45 Join parafin [0] (
16:39:45 Join kugel [0] (i=kugel@rockbox/developer/kugel)
16:39:45 Join T44 [0] (
16:39:45 Join Kaspar [0] (n=kaspar@
16:39:45 Join watto [0] (n=watto@
16:39:45 Join FOAD [0] (n=dok@
16:39:45 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
16:39:45 Join Llorean [0] (n=DarkkOne@rockbox/user/Llorean)
16:39:45 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
16:39:45 Join mc2739 [0] (n=mc2739@rockbox/developer/mc2739)
16:39:45 Join Omlet [0] (
16:39:45 Join hd [0] (n=jd@Wikipedia/HellDragon)
16:39:45 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
16:39:45 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
16:39:45 Join flydutch [0] (
16:39:45 Join petur [50] (n=petur@rockbox/developer/petur)
16:39:45 Join maruk [0] (n=papier@
16:39:45 Join Rob2222 [0] (
16:39:45 Join grndslm [0] (n=grndslm@
16:39:45 Join markun [50] (n=markun@rockbox/developer/markun)
16:39:45 Join Overand [0] (
16:39:45 Join fish_ [0] (
16:39:45 Join knittl [0] (n=knittl@unaffiliated/knittl)
16:39:45 Join rjg [0] (
16:39:45 Join hatseflats [0] (n=hatsefla@
16:39:45 Join ps-auxw [0] (
16:39:45 Join bzed [0] (
16:39:45 Join Dhraakellian [0] (
16:39:45 Join Zambezi [0] (i=Zulu@
16:39:45 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
16:39:45 Join BlakeJohnson86 [0] (
16:39:45 Join sbhsu [0] (
16:39:45 Join kadoban_ [0] (
16:39:45 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
16:39:45 Join tha [0] (
16:39:45 Join xavieran [0] (
16:39:45 Join FlynDice [0] (
16:39:45 Join Tomis [0] (n=Tomis@
16:39:45 Join matsl [0] (
16:39:45 Join bluebroth3r [0] (n=dom@rockbox/developer/bluebrother)
16:39:45 Join Zarggg [0] (
16:39:45 Join goffa [0] (n=goffa@
16:39:45 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
16:39:45 Join @ChanServ [0] (ChanServ@services.)
16:39:45 Join ThomasAH [0] (
16:39:45 Join yosafbridge [0] (n=yosafbri@
16:39:45 Join Slasheri [0] (i=miipekk@rockbox/developer/Slasheri)
16:39:45 Join GodEater [0] (n=bibble@rockbox/staff/GodEater)
16:39:45 Join jon-kha [0] (
16:39:45 Join lostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
16:39:45 Join rasher [50] (n=rasher@rockbox/developer/rasher)
16:39:45 Join dionoea [0] (
16:39:45 Join Xerion [0] (
16:39:45 Join jordan` [0] (n=jordan@
16:39:45 Join shodanX [0] (
16:39:45 Join jasio [0] (
16:39:45 Join Hadaka [0] (
16:39:45 Join aevin [0] (i=eivindsy@unaffiliated/aevin)
16:39:45 Join r00s [0] (
16:39:45 Join thegeek [0] (n=nnscript@
16:39:45 Join HBK [0] (n=hbk@
16:39:45 Join J-23 [0] (
16:39:45 Join evilnick [0] (i=4571af51@rockbox/staff/evilnick)
16:39:45 Join n17ikh [0] (n=n17ikh@
16:39:45 Join Lucky7hirteen [0] (n=freaking@
16:39:45 Join fxb [0] (
16:39:45 Join fyrestorm [0] (
16:39:45 Join amiconn [0] (i=quassel@rockbox/developer/amiconn)
16:39:45 Join pixelma [0] (i=quassel@rockbox/staff/pixelma)
16:39:45 Join Sajber^ [0] (
16:39:45 Join solexx_ [0] (
16:39:45 Join avacore [0] (i=nobody@
16:39:45 Join TaZzZ [0] (
16:39:45 Join alexbobp [0] (n=alex@
16:39:45 Join tarbo [0] (n=me@unaffiliated/tarbo)
16:39:45 Join mikroflops [0] (
16:39:45 Join killan [0] (
16:39:45 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
16:39:45 Join SIGSEGV [0] (n=user@
16:39:45 Join Utchybann [0] (
16:39:45 Join gtkspert_ [0] (
16:39:45 Join niekie_ [0] (
16:39:45 Join linuxguy3 [0] (
16:39:45 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
16:39:45 Join Torne [0] (i=torne@rockbox/developer/Torne)
16:39:45 Join jfc^3 [0] (
16:39:45 Join tmzt [0] (
16:39:45 Join Res1 [0] (
16:39:45 Join bughunter2 [0] (n=bughunte@unaffiliated/bughunter2)
16:39:45 Join YPSY [0] (
16:39:45 Join blithe [0] (
16:39:45 Join ehntoo [0] (
16:39:45 Join rvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
16:39:45 Join Galois [0] (
16:39:45 Join scorche|sh [50] (n=scorche@rockbox/administrator/scorche)
16:39:45 Join z35 [0] (
16:39:45 Join krazykit [0] (
16:39:45 Join jds2001 [0] (n=jds2001@fedora/jds2001)
16:39:45 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
16:39:45 Join zu_ [0] (
16:39:45 Join crashd_ [0] (
16:39:45 Join maraz [0] (
16:39:45 Join topik [0] (
16:39:45 Join chaos [0] (n=chaos@gentoo/user/ch4os)
16:39:45 Join preglow [0] (
16:39:45 Join lyngaas [0] (
16:39:45 Join Trista281 [0] (
16:39:45 Join CIA-80 [0] (n=CIA@
16:39:45 Join rphillips [0] (
16:39:45 Join elcan [0] (
16:39:45 Join crwl [0] (
16:39:45 Join B4gder [241] (n=daniel@rockbox/developer/bagder)
16:39:45 Join Kohlrabi [0] (
16:39:45 Join Tuplanolla [0] (n=jani@unaffiliated/tuplanolla)
16:39:45 Join Unhelpful [0] (n=quassel@rockbox/developer/Unhelpful)
16:39:45 Join jvd [0] (n=syscrash@poipu/developer/syscrash)
16:39:45 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
16:40:32geertvdijknvm, i just realized that i'll be notified by email when someone responds to the flyspray task so that has the same effect
16:40:32 Join webguest35 [0] (
16:41:04pixelmaif you don't uncheck the "notify when task changes"
16:41:13geertvdijkI don't :)
16:42:03CtcpVersion from freenode-connect!freenode@freenode/bot/connect
16:45:18webguest35hey guys what is the use of rockbox in my ipod nano?
16:46:20linuxstbwebguest35: (linked to from the front page of our website)
16:50:07 Join pamaury [0] (n=pamaury@
16:56:39geertvdijkany dev: any values beside VID/PIDs with and without HID enabled needed?
16:57:53gevaertsgeertvdijk: if possible also the serial numbers, and the same values for the OF
16:59:24geertvdijkserial numbrs?
17:02:02geertvdijkcould that be 0x03?
17:02:27geertvdijkseems like a short serial number to me,but it is wat USB View (microsoft tool) says after iSerialNumber:
17:03:11gevaertsoh, that's the index of the serial number in the string table then
17:04:06*gevaerts isn't too familiar with USB View
17:04:37geertvdijkno, me neiter, but the only other tool I have swamps me with information but no serial number. its more a snooper than just something that displays static device info
17:05:05geertvdijkgevaerts: what kind of value am I looking for
17:05:38gevaertssomething like 90000000000000000A4B4169405EB1BFD
17:06:13geertvdijk1000000000000000093B367880AE4FFF4 could be it?
17:07:04geertvdijkah, I've figured USBView out. For some reason it doesn't display all the necessary information unless you tell it to
17:07:52geertvdijkin the Options menu, enable Config Descriptors, then refresh (even if you've set it to autorefresh in config), and then you get more info where available
17:08:17geertvdijkmight anybody care xD
17:08:49 Quit gevaerts (Nick collision from services.)
17:08:59 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
17:09:36geertvdijksome more confusion, I'm getting 2 VID/PID pairs here?
17:10:31geertvdijkI've got idVendor which reads "0x0781", which is sandisk, and I've got iManufactorer which reads "0x01", which is rockbox.
17:10:41geertvdijksame for productid
17:10:49gevaertsiManufacturer is a string index
17:10:53gevaertsnot a PID
17:10:59gevaerts(or VID(
17:11:04geertvdijkah, that explains a lot, thanks
17:11:31geertvdijkis iSerialNumber a string index too then?
17:11:48Torneall these fields are fixed size
17:11:55kugelsomething is fishy with the theme site
17:12:26kugelunder h100 you can find various color themes (including one that's for a bigger resolution also)
17:14:35geertvdijkand for the OF, get all the values for both MTP and MSC or just MSC?
17:17:45 Quit DerPapst ("Leaving.")
17:22:07 Join phanboy4 [0] (n=benji@
17:25:10 Nick niekie_ is now known as niekie (
17:27:03geertvdijkthis is getting very weird, it's working for me now
17:27:21KasparI've been having connection problems with HID enabled on Ubuntu 9.04 ever since 3.4 was released...
17:27:38Kasparhow do I do the same checks under linux?
17:28:14geertvdijkKaspar: I spotted a tool for that seconds ago, let me check
17:28:59geertvdijkthere's an USBView for linux-tool on that page
17:29:09geertvdijkhavent used it myself yet though
17:29:33gevaertsKaspar: lsusb can do that, but I'm pretty sure this is a windows issue, not a rockbox one
17:31:12Kasparhm probablyit's just my PC... strange things usb have been happening every now and then
17:31:39 Join MethoS- [0] (n=clemens@
17:32:05Kasparbut ubuntu did never recognise the e200 as a MSP device since HID has been enabled
17:32:17KasparHID works
17:34:26geertvdijkit works for me, now, after booting into OF and changing usb mode to MSC
17:35:20geertvdijkI hit enter by accident, sorry abotu that. Anyway, I rebooted to rockbox after that, and it works now. To check out I rebooted into the OF again and set it to MTP, rebooted to rockbox and it still works. very odd.
17:36:00geertvdijkmy sansa wasnt connected during the rebooting by the way
17:37:15 Join funman [0] (n=fun@rockbox/developer/funman)
17:37:19Kaspareither it just quietly doesn't work or I get org.freedesktop.Hal.Device.UnknownError
17:37:34funmanKaspar: before working on clip+ we need to wait for the first firmware upgrade from SanDisk
17:38:47 Part Kaspar
17:46:48 Quit matsl (Read error: 110 (Connection timed out))
17:52:48funmanwebguest35: please do not write in caps, and i think this question is not related to rockbox
17:53:02 Join StealthyXIIGer [0] (n=stealthy@
17:53:08mc2739webguest35: that is off topic for this channel
17:55:48 Join Grahack [0] (
17:58:11 Quit webguest35 ("CGI:IRC")
18:02:55pixelmakugel: colour themes will work on the H100 unless they are using viewports - the one with the wrong dimensions won't and it appears incorrectly for some other screens too
18:02:55 Quit petur ("work->home")
18:05:04 Join TheSeven [0] (n=theseven@rockbox/developer/TheSeven)
18:05:19pixelmaif the theme is made for the correct dimensions of course and since there are targets with 160x128 colour displays, that is quite possible. They can possibly work on the monochrome M:Robe100 display if they don't use album art or backdrops
18:08:24 Join solexx [0] (
18:09:24 Quit solexx_ (Read error: 60 (Operation timed out))
18:11:01kugelah ok, i wasn't aware of that
18:12:16 Quit pamaury ("exit(*(int *)0 / 0);")
18:16:45 Join n1s [0] (n=n1s@rockbox/developer/n1s)
18:19:03TheSevenUtchybann: did you work out a function-based solution?
18:24:04kugelTheSeven: what is the exact problem? I read gcc is optimizing this away?
18:28:07 Join Xerion_ [0] (i=xerion@
18:29:42 Join Tomis2 [0] (n=Tomis@
18:29:53UtchybannTheSeven: sorry. no time to work on rockbox those days.
18:30:33 Quit Xerion (Read error: 60 (Operation timed out))
18:30:33 Nick Xerion_ is now known as Xerion (i=xerion@
18:31:05Utchybannkugel: my solution is an ugly macro that is not optimize by gcc. But we can't be sure that never version of gcc will not optimize it.
18:31:22kugelwhat solution?
18:31:57UtchybannI will open a patch in FS with the current macro. USEC_TIMER is bogus on nano2g.
18:33:48 Quit maruk ("Leaving.")
18:38:31 Quit Tomis2 (Read error: 60 (Operation timed out))
18:39:39***Saving seen data "./dancer.seen"
18:45:33 Quit Tomis (Read error: 110 (Connection timed out))
18:46:56 Join Tomis [0] (n=Tomis@
18:51:31TheSevenkugel: the nano2g usec timer is actually a 64bit 5usec timer. rockbox only uses the lower 32 bits, but these only seem to get updated when the high part is read (probably for consistency), so we need to fool gcc into reading something entirely unused
18:51:48TheSeventhe proper solution for this is converting USEC_TIMER to a function
18:52:42 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
18:55:07kugelTheSeven: or do it in assembly
18:55:19kugelI remember a similar problem
18:55:39TheSevenhow could i use assembly there?
18:56:06TheSeventhis is just a #define USEC_TIMER *((volatile uint32_t*)0xwhatever)
18:56:56kugelI think you need to read a part which is unused later
18:57:02TheSeveni would suggest #define usec_timer() *((volatile uint32_t*)0xwhatever) for the other targets and inline void usec_timer(void) { ... } for nano2g
18:57:36kugelUSEC_TIMER is not the unused part?
18:57:37TheSevenand of course s/USEC_TIMER/usec_timer()/ everywhere
18:57:49 Quit funman ("free(random());")
18:58:03TheSevenaccessing usec_timer needs to trigger reading a different reg *before* accessing the actually returned one
18:58:27JdGordonmornign all
18:58:38kugelright, and I was thinking to do that in inline assembly
18:58:39 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:59:07TheSevenour current hack to make it work is #define USEC_TIMER (((*(REG32_PTR_T)(0x3C700080))>>31)|1) * ((*(REG32_PTR_T)(0x3C700084)) * 5)
18:59:11TheSevenand I consider that just ugly
18:59:32TheSevenhow can you do a preprocessor macro that's returning something in asm?
19:00:53kugel{ long val; asm volatile ("ldr %0, =blaah\nldr %0, =read_addr", "=r"(val)); val }
19:01:09kugelI don't know the inline assembly syntax too well, but that should work
19:02:05kugelUnhelpful is an expert for that :)
19:02:14 Quit LambdaCalculus37 ("Leaving")
19:02:35kugelbut if that C code macro works, use that
19:02:53 Nick YPSY is now known as Ypsy (
19:03:01TheSevenwell, that c solution means that it isn't a macro any more on nano2g
19:11:46 Nick Ypsy is now known as YPSY (
19:13:12 Join bertrik [0] (
19:13:36 Nick YPSY is now known as Ypsy (
19:14:39geertvdijkcan anyone tell me if any requirements/targets have been set for 3.5?
19:15:13gevaertsas in "we must have X"? If so, no, we don't do that
19:15:25 Join liar [0] (n=liar@
19:16:47 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
19:16:48 Quit pixelma (Nick collision from services.)
19:17:06 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
19:17:08bertrikI think the condition is more like (date > 3 months since latest release)
19:17:17gevaertsbertrik: ==!
19:17:42gevaertsalthough we were an entire day late for 3.4 IIRC
19:17:48 Quit amiconn (Nick collision from services.)
19:17:49 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
19:18:09 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
19:21:23 Quit kugel (Read error: 110 (Connection timed out))
19:22:27UtchybannTheSeven: I have submit a patch (FS #10811).
19:23:33gevaertsTheSeven: 5usec or 5nsec?
19:23:56 Join AaronM [0] (
19:24:09TheSevengevaerts: usec, i think
19:24:31TheSevenit seems to tick every 250 pclk cycles
19:24:55gevaertsusec then
19:25:08gevaertsmc2739: does this e270r work with linux?
19:26:59 Join einhirn [0] (n=Miranda@
19:29:11 Quit StealthyXIIGer (Read error: 110 (Connection timed out))
19:29:57 Join robin0800 [0] (
19:31:48TheSevenhm, is it a known issue that if i connect usb while being in the wps, I will get the USB screen with all the wps texts that where scrolling, scrolling on top of the usb screen now?
19:31:55 Join dfkt_ [0] (i=dfkt@unaffiliated/dfkt)
19:33:21gevaertsI haven't heard that specifically, but I know there are places that don't handle connections properly
19:33:26LloreanTheSeven: Are you using a current build? I seem to recall a workaround going in, though maybe it was removed (there was some complaint)
19:34:20TheSevenno, the build is a couple of weeks old
19:34:44TheSeventhis seems to happen with basically everything that's scrolling
19:34:50TheSevenalso with too-long entries in the file browser
19:35:04TheSeveni'm in the process of updating to svn head right now
19:36:40 Join adiroiban1 [0] (n=adiroiba@
19:39:33TheSevenbah, someone broke audio!
19:40:35 Part Llorean
19:41:05 Quit T44 (Connection reset by peer)
19:41:09 Join T44 [0] (
19:41:57 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
19:42:13TheSevenok, that scrolling text bug was fixed
19:42:31TheSevenbut the build is totally unusable
19:42:38gevaertsso what do you prefer, working audio or working non-scrolling text? :)
19:42:59TheSeveni think i prefer to poke whoever broke this
19:43:49UtchybannTheSeven: r23722+FS #10811 works for me.
19:46:25TheSeven...and it just killed my FTL
19:46:35TheSevensomething is very very wrong with svn head
19:50:37TheSevenaha. as soon as i disable boosting it starts to work
19:52:01 Quit dfkt (Read error: 110 (Connection timed out))
19:52:16TheSevenoh no, not again
19:52:17 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
19:52:23TheSevenit's messing up PLL0PMS again
20:00:17 Quit einhirn (Read error: 131 (Connection reset by peer))
20:05:15 Join StealthyXIIGer [0] (n=stealthy@
20:10:03*LambdaCalculus37 is getting back to work on the GoGear SA9200 plugin keymaps
20:11:24 Join FOAD_ [0] (n=dok@
20:11:50 Quit dfkt_ (Read error: 110 (Connection timed out))
20:11:58 Join dfkt_ [0] (i=dfkt@unaffiliated/dfkt)
20:13:07TheSevenseems like the AHB doesn't like freq switching and decides to mess up memory accesses
20:15:28CIA-80New commit by theseven (r23747): Fix the S5L8701 µsec timer
20:15:29CIA-80New commit by theseven (r23748): Enable wheel acceleration for iPod Nano 2G
20:15:38 Quit LambdaCalculus37 ("Fwump")
20:19:21 Quit shodanX (
20:19:21 Quit rasher (
20:19:21 Quit dionoea (
20:19:21 Quit jordan` (
20:19:26 Join dionoea_ [0] (
20:19:28 Join Thundercloud [0] (
20:19:35 Join shodanX_ [0] (
20:19:52 Join rasher [0] (n=rasher@
20:20:01NJoinjordan` [0] (n=jordan@
20:20:06 Quit dfkt (Read error: 110 (Connection timed out))
20:20:25 Quit FOAD (Read error: 145 (Connection timed out))
20:20:25 Nick FOAD_ is now known as FOAD (n=dok@
20:20:33 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
20:21:22 Quit FOAD ("I'll be back")
20:21:35 Join FOAD [0] (
20:22:44 Quit gevaerts (Nick collision from services.)
20:22:53 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
20:25:08 Quit AaronM ("Emo Time In My Corner... //_-")
20:25:17 Quit JdGordon ("Leaving.")
20:27:18 Join Kopfgeldjaeger [0] (
20:31:20 Join domonoky1 [0] (
20:34:00 Join kugel [0] (n=kugel@rockbox/developer/kugel)
20:34:49*domonoky1 thinks he found out why the recording trigger menu is broken. Somehow the gui_sync_list or settings doesnt work any more with a setting with negative step ? If i change the step to 1, it works again, but ofcourse decrements instead of increments.
20:37:19TheSevennano2g rolo is also broken once again...
20:37:56 Quit BHSPitLappy (Read error: 145 (Connection timed out))
20:38:33kugelthe fix seems easier as thought now?
20:39:13kugelFIVE_USEC_TIMER is actually ONE_FIFTH_USEC_TIMER?
20:39:43***Saving seen data "./dancer.seen"
20:39:50*kugel still doesn't understand why this is a 64bit reg
20:40:02 Part adiroiban1
20:48:58TheSevenit obviously is one though
20:49:29TheSevenand yes, i found a way to keep it a c macro, that seems to work
20:49:43 Quit domonoky (Read error: 110 (Connection timed out))
20:50:19kugelso it just counts up till wrapping at 64bit?
20:50:24TheSevenprobably yet
20:50:36kugelI imagine one could just ignore the upper 32bit
20:50:52 Nick hd is now known as HellDragon (n=jd@Wikipedia/HellDragon)
20:50:53TheSevenone could, but one needs to read them nevertheless
20:51:06TheSeventhe lower 32 bits get updated when reading the upper 32 bits
20:51:40TheSevenwhat i don't get though is why writing to 0x3c500000 overwrites the data at 0x3c50004 with the same pattern with svn head, while it didn't do so with the version i had on my ipod until today
20:51:55TheSevenin other words: writing CLKCON trashes PLL0PMS, which is quite bad
20:52:02kugeldid you check the value of 32bit really counting up? maybe it's just a reg that needs to be read (for whatever reasons) but doesn't contain a useful value
20:52:25kugelof the upper*
20:52:34n1scan't you just read it as a 64 bit var in c?
20:53:07n1sor is gcc cleaver enough to not read the upper half if it isn't used?
20:53:13n1sclever, even
20:53:21kugelit must read it
20:53:30 Nick fxb is now known as fxb__ (
20:53:45domonoky1can somebody explain to me what the check in option_select.c around line 248 does ? it seems wrong to me and most likely is the cause for the recording trigger bug
20:54:50kugeln1s: a) 64bit introduces useless overhead, and b) the upper 32bit could be doing more harm than good if it's really contains only nonsense
20:55:25 Part watto
20:55:25TheSevenit's thrown away anyways
20:55:42n1skugel: i meant to throw it away after read
20:55:57 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
20:56:17kugelyes, it's *implictely* thrown away
20:56:49n1s*REG_FOO & 0xffffffff or something if REG_FOO is a 64 bit type
20:56:50TheSevenwe can't really read it while explicitly throwing it away...
20:57:13TheSeven(except using some asm hack or a function)
20:57:22amiconnTheSeven: You don't need to turn USEC_TIMER into a function to make gcc read the high part
20:57:59kugeldomonoky1: going by svn blame, it is to fix quickscreen going mad for negative setting values
20:58:04amiconnJust read it as volatile uint64_t and cast afterwards
20:58:35n1sthat's what i was trying to suggest!
20:58:38*TheSeven wonders if he actually wants to cast it... or just leave that to the code using the macro
20:58:38*kugel wonders if amiconn keeps track of svn commits :)
20:58:54amiconnAnd btw, this behaviour actually makes sense
20:58:55domonoky1kugel: but it causes all other menus with negativ steps to go insane. (only min and max available)
20:59:16kugelwhich for example=
20:59:21amiconnIf the parts would be readable independently, you would never be sure that you've read a correct value
21:00:03 Join pamaury [0] (n=pamaury@
21:00:09amiconnSince a 64 bit access is non-atomic, the counter may advance between the two 32 bit reads
21:00:26pixelmakugel: domonoky1 mentioned the triggered recording settings
21:01:14kugelthat's not really a usual menu
21:01:19domonoky1kugel: for example the recording trigger thresholds are completly broken at moment.
21:01:44domonoky1kugel: but it uses the normal menu system.. so it should affect all menus with negativ steps
21:02:54amiconnkugel: That commit isn't what I mean
21:03:33amiconnOr rather, doesn't do what I mean
21:05:28 Join pixelma_ [0] (
21:05:31 Join panni_ [0] (
21:05:35 Quit pixelma (Nick collision from services.)
21:05:41 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
21:05:45amiconnHmm, the two parts are in the opposite order of what I'd expect on a little endian system...
21:06:35domonoky1ah, now i see the real problem, the INT_SETTING expects now the smallest number in max and the biggest in min, for negativ steps..
21:06:52TheSevenyes, that's the problem
21:07:00n1sdomonoky1: that sounds backwards :)
21:07:09TheSevenelse i would have used uint64_*
21:07:13 Nick dfkt_ is now known as dfkt (i=dfkt@unaffiliated/dfkt)
21:07:13 Join dionoea [0] (
21:07:27 Join jds2001_ [0] (n=jds2001@fedora/jds2001)
21:07:41 Quit goffa (Remote closed the connection)
21:07:46 Quit jds2001 (Remote closed the connection)
21:07:50 Quit dionoea_ (Remote closed the connection)
21:07:54amiconnTheSeven: That would still be possible if you don't care about the 64 bit value being correct
21:08:04 Quit Trista281 (Broken pipe)
21:08:15TheSevenwell, i care about the lower 32 bits being correct
21:08:19 Join _Tristan_ [0] (i=tristan@
21:08:20amiconnIt would probably result in the same asm sequence as the currrent svn version
21:09:02amiconnYou could also use the comma operator, like this:
21:09:58n1sTheSeven: why not just *uint64_t >> 32 ? am i missing something?
21:10:15amiconn #define USEC_TIMER (*(REG32_PTR_T)(0x3C700080), (*(REG32_PTR_T)(0x3C700084)) * 5)
21:11:26TheSeventhe current way allows us to use it as both a 32bit and 64 bit field, depending on what you do with the result value
21:11:56amiconnI'm not talking about changing FIVE_USEC_TIMER anymore
21:12:18TheSevenyes, but the same is true for USEC_TIMER
21:12:27amiconnUSEC_TIMER is 32 bit
21:12:38amiconnHmm, or not
21:12:44 Join goffa [0] (n=goffa@
21:12:51TheSevenUSEC_TIMER is whatever type you use it as
21:13:05amiconnAtm it seems it's not. You might save some code if you make it 32 bit again
21:13:34*TheSeven would really like to know why PLL0PMS gets corrupted with his current build
21:13:41kugelis gcc smart enough to use only 1 register for that comma expression?
21:13:42amiconnHmm, and I'd suggest parentheses. Right now USEC_TIMER isn't safe
21:15:15*TheSeven thought it had parantheses
21:15:46TheSevenah, only 5µsec
21:15:47amiconnIt did before r23747
21:16:03amiconnkugel: Try it...
21:16:18CIA-80New commit by theseven (r23749): Add missing parentheses to S5L8701 USEC_TIMER
21:18:13kugelif only the sansas had the trouble with a too accurate dedicated usec timer :/
21:18:55kugelinstead, we have only 2 timers and those even clock with stupid 1.5MHz
21:19:27CIA-80New commit by Domonoky (r23750): make recording trigger threshold settings work again.
21:23:39*domonoky1 slaps kugel for the samsa.
21:23:59TheSevena mistyped typo...
21:24:34kugelyea, fail!
21:27:37*kugel desperately tries to get an interrupt going properly on the mini2440
21:29:12mc2739gevaerts: I do not use the on the linux system, but when I connected it to get the lsusb data I was also able to see the disk partitions.
21:29:16kugelthe otherwise great datasheet is a bit confusing when it comes to external interrupts
21:30:01gevaertsmc2739: ok, thanks
21:33:17 Join kugel__ [0] (
21:33:32 Quit kugel (Nick collision from services.)
21:33:39 Nick kugel__ is now known as kugel (
21:35:17*TheSeven wonders if the midi player is broken only on nano2g...
21:36:48geertvdijktheseven: which revision?
21:36:56TheSevensvn head
21:37:07geertvdijkI'll check on my sansa :)
21:37:24geertvdijkr23750-091125 right?
21:37:49 Quit HBK ()
21:38:49 Join froggyman [0] (
21:39:04TheSevenr23748 to be exact
21:40:43 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
21:41:13pixelmamidiplayer doesn't run that well on arm targets, not sure what you exactly mean with broken (I admit I haven't tried in a while)
21:41:22geertvdijkoh. I just got 23750 on now
21:41:34TheSevenit's just playing static
21:42:06liarTheSeven: whats wrong with r23748?
21:42:22geertvdijkTheSeven: I'm getting the patchset now and I'll tell you what my sansa does right after
21:42:24 Join FOAD_ [0] (
21:43:00pixelmaTheSeven: midi works okish on my c200 with r23718
21:43:18pixelmaa v1 c200
21:45:15TheSevenon my nano it's playing something in between white and pink noise :-P
21:45:23geertvdijkpink noise?
21:45:31geertvdijkare you sure there are no unicorns around?
21:45:46pixelmaTheSeven: tried a different midi?
21:46:01TheSevenresults in slightly different noise
21:46:04kugelcolorful sounds? must be an awesome experience
21:46:10pixelmanice ;)
21:46:22geertvdijkr23750, 12 midi tracks in the file, no white noise, just music
21:46:34geertvdijksansa e250
21:46:39gevaertskugel: it's an iaudio device :)
21:47:52pixelmaTheSeven: is that with the patchset available through the wiki?
21:50:38 Quit Thundercloud (Remote closed the connection)
21:51:14TheSeventhe static even continues to play after i quit the midi player
21:51:39TheSevenor rather it switches to a differend kind of static
21:53:01 Join HBK [0] (
21:53:56 Quit FOAD (Read error: 110 (Connection timed out))
21:53:57 Nick FOAD_ is now known as FOAD (
21:55:50domonoky1kugel: ah i think i now understand your code. so the problem is, that the interrupt only happens once ?
21:57:01 Join dfkt_ [0] (i=dfkt@unaffiliated/dfkt)
21:57:13kugelyes, at boot where the sd is untouched. it doesn't fire for any insertion after
21:57:29 Quit dfkt (Nick collision from services.)
21:57:32 Nick dfkt_ is now known as dfkt (i=dfkt@unaffiliated/dfkt)
21:57:55kugelvery early at boot, presumably on or shorly after the first card_enable_monitoring_target() call
21:58:59domonoky1thats strange.. why should it see a rising edge on this pin early on startup ? or does some other code maybe modify this pin ?
21:59:29 Join FOAD_ [0] (
21:59:48kugelnothing is writing to it
22:00:18 Quit FOAD (Read error: 60 (Operation timed out))
22:00:19 Nick FOAD_ is now known as FOAD (
22:02:41domonoky1system_init() changes GPGUP (pullup) but that should be before monitoring is enabled
22:03:57n1sTheSeven: midiplayer uses 22kHz samplerate if it can so maybe something regarding that is broken on the nano2g (it isn't used in many places9
22:03:58kugelhm, strange things happen when disabling irq during card_enable_monitoring_target()
22:05:24domonoky1button_init also seems to enable a pullup on GPG9
22:06:38domonoky1oh SD detect is on gpg8, so ignore me :-)
22:07:21kugelgpg4 actually
22:07:49kugelI don't quite understand the naming, but GPG4 means GPGDAT pins 8 and 9
22:09:10TheSevenn1s: this should in theory work, but is completely untested
22:09:21TheSeventhanks for the hint... i bet it's somewhere around there
22:09:32kugelmaybe I confuse somethinghere
22:10:17n1sTheSeven: since the cpu in the nano2g is pretty fast it might be a good idea to force 44.1kHz anyway as that sounds better
22:10:21*kugel gives GPG8 a try
22:11:01TheSevenwe should fix the root cause for the trouble though
22:11:42geertvdijkI'm wondering: is there a MIDI-api of some sorts in rockbox? For lua or c?
22:13:18n1sTheSeven: iirc, there's a test_sampr plugin that might be usefull
22:13:43amiconngeertvdijk: No, since the core doesn't handle midi, and there can only be one plugin running at a time
22:13:45TheSeveniirc that was one of the plugins that didn't even start
22:14:13amiconnn1s: Doesn't doom also use lower sample rate?
22:14:22geertvdijkamiconn: so right now, if I want to make some plugin that makes use of MIDI I'd have to copy the MIDI-plugin and go from there?
22:14:52kugeldomonoky1: ok, I was being confused by the ds, it works now
22:14:58amiconnYes, or write your own midi lib ;)
22:15:02n1samiconn: i think so
22:15:08kugelbut the isr runs twice
22:15:22*TheSeven cant remember hearing any sound in doom
22:15:24geertvdijkamiconn: fair enough, but I think I'll be way in over my head if I'm going to do that :P
22:15:32amiconnSeriously, if this would be useful for other plugins, the midi part of the midiplay plugin should be put into the pluginlib at least
22:15:35n1sgeertvdijk: there is a disabled, old, bitrotted plugin called beatbox that uses midiplayer as alibrary
22:15:41TheSevenbut i also can't remember having ever plugged speakers while trying doom
22:16:19n1samiconn: it's far too slow o
22:16:25n1sto be usefull for much
22:16:45amiconnOptimization should help there
22:16:53domonoky1kugel: nice, is it run twice directly one after the other ?
22:17:01geertvdijkn1s: I have looked at it once or twice, but couldn't get anything done with it to be honest
22:17:34n1si tried and it is actually quite abit faster than it was when i started but i have hit a roadblock and think the main synth loop needs to be entirely rewritten
22:17:40geertvdijkamiconn: agreed, but I really am not a good programmer. should I make a flyspray ticket to suggest this?
22:18:07n1sgeertvdijk: flyspray is not for suggestions
22:18:10kugelacking the irq is the first thing I do
22:18:33geertvdijkn1s: my bad. I meant the forums ofc
22:18:41domonoky1kugel: you could try with the filter option of eint. and maybe ack the irq as last thing in the isr
22:18:53TheSevenno sound in doom at all
22:19:26kugelwow, the other code seems to work at the first try (the other part is to detect changed binary as it does for usb)
22:20:10*bluebroth3r comes in again ... after around 22 hours of RL
22:20:22 Nick bluebroth3r is now known as bluebrother (n=dom@rockbox/developer/bluebrother)
22:20:35kugelthe uart_printf seems to be too slow to make the ack in time
22:21:12bluebrotherdomonoky1: did you add italian to the rbutil translation status table?
22:21:13domonoky1yes, uart_printf is very slow.. but you could also use logf over serial :-)
22:21:36TheSevenis this faster?
22:21:38 Join stripwax [0] (
22:21:45kugelhow is that faster than uart_printf over serial?
22:21:57TheSeventhe usb logf is buffered, but i can't remember the plain serial one being buffered...
22:22:05domonoky1bluebrother: nope
22:22:26kugelTheSeven: if debug messages where buffered they would be pretty useless
22:22:35TheSevennot always
22:22:39domonoky1ah, if logf over usb doesnt buffer, then it wont help..
22:22:52kugelTheSeven: often enough
22:22:58domonoky1but i though it processed it in another thread.
22:23:09bluebrotherhmm. Why does foswiki doesn't show the authors of changes anymore? Or am I missing them?
22:23:15TheSevendomonoky1: the usb logf, yes
22:23:39*kugel rolos rockbox after inserting a SD with a newer rockbox binary :>
22:24:34*TheSeven grumbles about a rolo that fails more often than not
22:24:44AlexPbluebrother: A good point - history shows them, but not an individual revision it seems
22:24:46kugelbluebrother: works in the history view
22:25:29bluebrotherAlexP: ah &minus;&minus; I usually use the direct revision comparison.
22:25:47bluebrotherbut it also seems to show the wrong diff &minus;&minus; for example see;rev2=73
22:26:06bluebrotherit shows a change in the Chinese line but it isn't changed at all. Instead the following line was added.
22:26:09pamauryI have a small question: is multivolume support actually target specific or a global (config) setting ?
22:26:35domonoky1bluebrother: nope, i removed a whitespace in this revision..
22:26:57bluebrotherdomonoky1: ah, that explains things :)
22:27:17kugeldomonoky1: I assume that would be useful for other targets too (cowond2, onda)
22:28:12*bluebrother should clean up his langstat script and commit it
22:28:18domonoky1kugel: yes, it might also be good for those..
22:29:26kugelbtw, I was able to replace all uart_printfs in the sd driver by delays and still have rockbox boot
22:29:35kugeland it's even booting faster :p
22:31:00domonoky1kugel: i can remove all uart_printf and all delays, but the last one in send_command, and it still works and is even more faster :-)
22:31:21kugelwhy is that not in svn then?
22:31:52domonoky1because i wanted to find out why this last delay is needed, but didnt found the time for it..
22:33:00domonoky1but feel free to commit this if you want.
22:33:53domonoky1i tested it with the test_disk plugin, so it surely is fine with this last delay in place.
22:33:55 Quit panni_ (Read error: 113 (No route to host))
22:36:11 Join petur [50] (n=petur@rockbox/developer/petur)
22:39:46***Saving seen data "./dancer.seen"
22:42:24 Join dfkt_ [0] (n=dfkt@unaffiliated/dfkt)
22:44:05 Quit HBK (Read error: 60 (Operation timed out))
22:45:18 Join panni_ [0] (i=hannes@
22:46:56 Quit Grahack ("Leaving.")
22:47:07 Join many [0] (
22:47:33kugelnice side-effect: settings are saved upon rolo :)
22:48:15 Quit dfkt (Read error: 60 (Operation timed out))
22:49:22 Join HBK [0] (
22:49:35 Quit many (Client Quit)
22:52:57pamaurygevaerts: I don't know if you or not but I asked and nobody answered :( is multivolume support actually target specific or a global (config) setting ? That's because I'm implementing multivolume support for MTP and I don't know how/where to enable it (config.h ?)
22:53:27gevaertspamaury: what player do you use?
22:53:52pamauryI only have a e200
22:54:25pamauryShould I set it config-e200.h ?
22:54:45gevaertsthat one has multivolume automatically
22:55:03gevaertsconfig.h sets HAVE_MULTIVOLUME if HAVE_MULTIDRIVE is defined.
22:55:38gevaertsYou can do it by hand, but that's only required if you have a single-drive player and you want multiple partitions
22:55:47pamauryOk, thanks.
22:58:49*Unhelpful is an expert now?!
23:00:40Unhelpfulkugel: i'm not some kind of expert, and i have no idea what you're trying to do with that asm? load a value into a register, and then load another value over it? :P
23:01:13kugeljust forget about that :P
23:01:30kugelproblem is solved in the meantime
23:08:26 Join wodz [0] (
23:09:53 Quit n1s ("Lmnar")
23:12:03 Join Stephen [0] (
23:12:31 Nick Stephen is now known as Guest2556 (
23:12:41 Nick Guest2556 is now known as Stephen___ (
23:13:22 Join Thundercloud [0] (
23:14:06Stephen___when uploading a theme can you not have 2 or 3 variations in the 1 zip ?
23:14:48Unhelpfulthere was really nothing wrong with the macro imo
23:15:17 Quit dfkt_ ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
23:15:23Unhelpfulpamaury: you're implementing MTP?
23:15:44pamaury(at least trying to do so :))
23:15:55Unhelpfuloh, sweet. not that there's a linux client for it that works worth a damn. :/
23:16:21wodzhello, I am working on port rockbox to mpio HD200. Currently I look at backlight routine in OF. It looks like the sequence is as follows: GPIOx low, delay - turns backlight off; GPIOx low, delay 1ms, repeat n times (GPIOx high, GPIOx low), GPIOx high. The number of pulses sets brightnes of backlight. Have anyone seen such scheme of controlling backlight? I am asking because I am unable to reproduce this 'by hand' If I use OF routines to toggle GPIO it works
23:16:31Stephen___itll be handy for me as all usb is blocked except mtp. good work i say pamaury
23:17:11 Join takeshita_kenji [0] (n=takeshit@
23:17:14bertrikwodz, I've seen such mechanism in some other OFs too
23:17:59pamaurygevaerts: (but of everyone interest), if you want to have a look at the mtp code, I have changed the repository I'm working on, it's now at pamaury/rockbox/commits/mtp-new">, I think I have reduced all the collateral changes to the minimum (there were some useless modifications before).
23:17:59wodzbertrik: Is it used to control some led driver?
23:18:23bertrikI'm still confused by that. The samsung yp-s3 OF does something like that IIRC, but if I just PWM it, it also dims
23:18:25takeshita_kenjiI was reading up on the status of 2G iPod Nano support, and came across iLoader. Is it applicable to the 6G iPod Classic, too?
23:18:33Stephen___how far along would you say you are pamaury ?
23:18:56TheSeventakeshita_kenji: no, just nano2g
23:19:02bertrikhowever, the backlight turns off at some point when going to low brightness
23:19:33bertrikwodz, I haven't found any datasheet or anything like that that describes the mechanism, I've only seen it in OFs
23:19:39Stephen___phew just uploaded 15 themes to the theme site. :)
23:19:48wodzbertrik: what You mean by 'PWM it'?
23:20:11kugelUnhelpful: my macro?
23:20:14takeshita_kenjiThat brings me to the next question: If I wanted to get a new portable audio player that works with RockBox, what should I get?
23:21:06bertrikapply a signal with variable duty cycle to the backlight pin and a frequency around a few 100 Hz
23:21:39wodzbertrik: aa software PWM I see
23:22:01kugelbertrik: have you had a chance to test the buffering patch on the c200v2?
23:22:06pamauryWell it depends from the point of view. Basically, the device is correctly detected under windows and with libmtp (I have not tested with MacOs since the major introduction of MS OS descriptors). Under windows, there are still lots of unsolved problems like the fact that that at some point the device doesn't get request anymore and the fact that WMP doesn't detect it. But with libmtp I can interact with, even mount with mtpfs, browse it, modify/add/remo
23:22:06pamauryve things, etc...
23:22:13bertrikwodz, built-in timers can often do something like that, you program it for a cycle of say 10 ms, and make the timer only turn its output on for 5 ms to achieve 50% dimming
23:22:22 Quit bmbl ("Bye!")
23:22:45bertrikkugel, I haven't tried any patch in the past few days
23:23:20wodzbertrik: ok but in this case setting GPIO high should turn backlight full on
23:23:28kugelUnhelpful: what to do about the F/X eabi build failure?
23:25:12 Part froggyman
23:25:26takeshita_kenjiDoes anyone know a good recent portable audio player that works with RockBox?
23:26:16pamauryStephen___: But there are lots of missing feature like multiple volumes, like metadata for musics, like playlists support/edition/... and there are some requests that should be implemented but there are redundant with some already implemented (except that there are more efficient). Anyway, I'm optimistic because the underlying protocol seems stable so now the focus is really on features.
23:26:36bertrikwodz, the ones I've seen so far can be turned fully on by just making the backlight pin high, doesn't that work for you?
23:26:49wodzbetrik: no
23:27:39Stephen___cool good work pamaury would love to see that in rockbox. 'll be watching you work. :)
23:28:39 Quit Omlet ("( :: NoNameScript 4.22 :: )")
23:29:16bertrikwodz, amiconn might know this
23:29:42bertrikmaybe the timing of the pulses is very critical
23:30:17amiconnI guess it's a similar backlight controller as the one used in iPod Video and Nano G1
23:30:27bertrikor maybe you have to turn on something else too, to make backlight work, like some kind of high-voltage generator (often something like 15V is used to drive a bunch of LEDs in series)
23:30:39amiconnThis can be dimmed up and down (with short/ long pulses
23:31:22amiconnBut these ipods have two separate gpio pins for backlight power and dimming
23:31:49takeshita_kenjiNobody knows?
23:32:33 Join ameyer [0] (
23:32:37wodztakeshita_kenji: I think there is no recent DAP which can be rockboxed
23:33:06takeshita_kenjiOh well. Thanks anyway.
23:33:09 Quit takeshita_kenji ("Bye.")
23:34:05wodzbertrik: good point - I will measure voltage on backlight diodes
23:34:27kugelwodz: yes, I have seen this scheme :)
23:34:42kugelI have implemented (with the help of some other guys) in rockbox
23:34:47*TheSeven likes the nano2g backlight where one just has 3 i2c regs: on/off, brightness and dimming speed. the rest (dimming etc.) is done in hardware, even when just setting a new brightness level
23:35:16wodzkugel: where can I look at this code?
23:35:28kugelsoftware backlight fading that is, it basically simulates PWM fading in software
23:36:10kugelis the port at a state already where you can worry about backlight fading?
23:36:39kugelwodz: I hope that page can help you
23:37:15wodzkugel: It is not problem of backlight fading but problem how to turn it on
23:38:32wodzsimply setting GPIO high is not enough
23:38:40kugelfor plain backlight you just need to implement _backlight_off() and _backlight_on()
23:39:50wodzkugel: You are talking on rockbox software level, I am asking If anyone seen such scheme of turning on backlight at hardware level
23:39:54 Quit ameyer (Remote closed the connection)
23:40:07kugelyou probably want to look at PWM yes, that theme doesn't look like what software fading was designed for
23:40:21stripwaxbug FS #10395 . is it just me, or would deleting lines 647-649 of apps/playlist_viewer.c just fix that? [seems to .. ]
23:40:22domonoky1wodz: and the OF does enable the backlight with just one gpio line ?
23:41:09stripwaxThe lines in question say "deselect track move if playlist changes while in the playlist viewer" but .. why ? remove the code and the bug is fixed .. ? anyone know why the deselect logic is there at all?
23:41:49wodzdomonoky1: I described what OF do to turn on backlight a bit earlier
23:41:53kugelI guess because inserting was/is buggy when the current track index changed or something similar?
23:42:14amiconnSvn history should tell
23:42:29kugelwodz: several targets do it that way I think, look at ipods in particular
23:43:13domonoky1wodz: yes, i saw it, but you only mentioned GPIOx, if it would use 2 gpio lines i could imagine that it would clock out some value to a led-driver
23:43:35kugelalthough it seems weird to find the "wait 1ms" in the code, that sounds more like a software implementation. usually the pwm hardware has its own timer
23:44:20kugelbut you can't even turn it on with setting the pin high?
23:45:27amiconnkugel: Not if the timing is used for communicating with the hw pwm controller
23:45:53amiconnI already pointed that out; check the brightness setting code for iPod Video + Nano G1
23:47:18amiconnI think this is the same thing, just via the same pin that also does the enabling
23:47:18wodzamiconn: Yes I look at backlight-nano_video.c now
23:47:18amiconnAnd if the pwm controller starts at the lowest level, just setting the pin to high level won't enable the light
23:47:33 Join Rondom [0] (
23:48:17CIA-80New commit by stripwax (r23751): Take out that code from the dynamic playlist viewer that aborts track move-in-progress on playlist change .. based on assumption it doesn't do ...
23:48:18amiconnIt's rather weird though. Does the OF allow changing brighness, and will it change brightness "live" when adjusting it?
23:48:35amiconn(i.e. without disabling the light before the new level becomes effective)
23:50:30wodzamiconn: looking at the function in OF i dont think so - It always drive GPIO low and make some delay before anything else.
23:54:36 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
23:54:38CIA-80New commit by kugel (r23752): Add HAVE_HOTSWAP_STORAGE_AS_MAIN to detect changed binary after re-inserting the storage on targets that run rockbox from hotswappable storage.
23:54:54 Nick fxb__ is now known as fxb (
23:55:20*amiconn is reminded of the early stages of the Ondio port...
23:56:14 Nick Ypsy is now known as YPSY (
23:56:35wodzamiconn: the thing I dont understand is that I do not see in OF any timed delays between driving pin low and high.
23:56:57FlynDiceI need a start point for investigating, can someone tell me where I am in the code or what are the sequence of events when I select a file and it starts playing. I'm hanging just after the wps screen appears...
23:57:00 Quit stripwax ("")
23:57:30amiconnPerhaps that chip is different, in that it only dims in one direction, and is able to understand even very short pulses?
23:57:45 Join stripwax [0] (
23:58:33kugelFlynDice: somewhere in tree.c
23:58:52FlynDicekugel: thx

Previous day | Next day