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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2009-01-09

00:01:19 Join solexx [0] (
00:02:59 Quit tessarakt (Read error: 110 (Connection timed out))
00:04:00linuxstbMarcGuay: Are you saying it still didn't work without the button_init code added?
00:04:31MarcGuayRAM is 4Mx16 SDRAM, which I interpreted as err... Forgot.... button_init code didn't change anything...
00:04:47MarcGuayWhere is RAM defined again? :)
00:05:17linuxstb4Mbit*16 should be 2MB
00:05:26 Join tessarakt [0] (
00:05:40MarcGuayOkay, I put 32. Smart!
00:05:45 Quit evilnick (" ajax IRC Client")
00:05:53MarcGuayIt's measured in MB in tool/config?
00:06:07linuxstbBut remove the button init code - I think that also breaks it...
00:07:13 Quit bertrik ("Leaving")
00:07:24linuxstbDue to the hard-coded "[pc, #-28] lines - which assume the code before it is a certain size. (badness by me...)
00:07:29*pixelma found a weird (to describe) bug, could be quite uncomfortable for someone with a target lacking hard power-off and using repeat...
00:08:04pixelmasomething for JdGordon|zzz when he wakes up.. I fail to come up with a nice summary to file a bug report
00:08:12MarcGuaylinuxstb: Gotcha. My use and understanding of mktccboot is correct though, right?
00:08:17 Quit itcheg (" ajax IRC Client")
00:08:33linuxstbMarcGuay: Yes, I think so.
00:09:59 Quit ender` (" I will remember that any vulnerabilities I have are to be revealed strictly on a need-to-know basis. I will also remember t")
00:11:12 Quit Unhelpful (Read error: 110 (Connection timed out))
00:13:09 Quit solexx_ (Read error: 110 (Connection timed out))
00:13:35 Quit Xerion (Read error: 54 (Connection reset by peer))
00:13:37MarcGuaylinuxstb: The player.rom created with TCCBOOT enabled should be useless until it's merged using mktccboot, correct?
00:14:22MarcGuayThe c100 makes a funny buzzing noise when you upload to it...
00:14:43linuxstbMarcGuay: Yes, I don't think it will work by itself.
00:15:03MarcGuayOkay... Now I'm getting no reaction from it... Crud..
00:15:21 Quit tessarakt ("Client exiting")
00:17:20linuxstbMarcGuay: After uploading something with tcctool?
00:17:44 Quit bluebrother ("leaving")
00:17:48linuxstbMarcGuay: BTW, I've just committed a fix to remove those #-28s - so the code should be more robust to you changing things.
00:17:55MarcGuaylinuxstb: Something borks when TCCBOOT is defined. If I upload the dualbootloader with tcctool I get nothing, hold switch in both positions.
00:18:30MarcGuayIf I undefined TCCBOOT I can display the logo/debug info.
00:19:03linuxstbYou mean the dual-boot bootloader merged with the OF?
00:19:39MarcGuaylinuxstb: Re: " If I upload the dualbootloader with tcctool I get nothing, hold switch in both positions." Yes.
00:20:10MarcGuayIt could be that it needs that button init before it can detect the hold switch.
00:20:15robin0800jhMikeS: thanks for updating the patch it appears to be working now
00:20:35linuxstbWhat file are you passing to mktccboot? bootloader.bin?
00:22:40 Quit Strife89 ("Dinner time! :)")
00:22:41MarcGuaylinuxstb: I'm passing player.rom. It seems the same as bootloader.bin, unless...
00:23:11linuxstbNope... It's not obvious, but you should use bootloader.bin
00:24:03MarcGuaylinuxstb:Using bootloader.bin lights up the LCD....
00:24:45MarcGuayNo OF, though... Err, hold switch in the other position lights it up as well. I'll test some more...
00:25:21linuxstbHave you merged the changes I just committed to crt0.S?
00:25:46MarcGuayRewind - that was using a non defined TCCBOOT with mktccboot. re: crt0.S - yes, and I'm about to test with the button_init...
00:27:06jhMikeSrobin0800: hey, np
00:27:35 Join denes [0] (
00:27:57MarcGuaylinuxstb: Interesting. Now the LCD flickers a bit as if it's stuck in a loop waiting for something. (TCCBOOT defined, mktccboot, button_init in crt0.S, hold switch in both pos.)
00:28:15jhMikeSrobin0800: if you're charging now, keep an eye on it and report if it stops properly (the usual).
00:28:26 Join ze [0] (i=ze@
00:29:07 Quit MethoS- (Remote closed the connection)
00:29:39MarcGuaylinuxstb: Excellent! Removing the button_init gets it booting into the OF. Dualbooting, AOK.
00:29:58MarcGuaySeems that -28 was the problem?
00:30:16MarcGuayOr perhaps me using player.rom instead of bootloader.bin.
00:31:03linuxstb-28, together with button_init was 1 problem. Using player.rom was another. Either would have both stopped it working (I think).
00:31:13 Quit flydutch ("/* empty */")
00:32:00linuxstbBut if you've merged my commit (removing the -28s), then it the button_init code shouldn't stop it working.
00:32:08robin0800jhMikeS: Spoke too soon it stopped at 60% and now only connects and won't charge
00:33:02 Quit gevaerts (Nick collision from services.)
00:33:11 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
00:33:29MarcGuaylinuxstb: Now to see if I can incorporate the NAND code... Any suggestions on where to start?
00:33:42 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
00:33:53jhMikeSdid you see an error in the battery screen?
00:34:01MarcGuaylinuxstb: It looks like adding the button_init code actually breaks button detection. It loads RB if the hold switch is in either position..
00:34:03linuxstbNo, the NAND code is a mystery to me. But shouldn't it just be included anyway for a TCC77x build? i.e. I wouldn't think there would be anything to do.
00:36:35MarcGuaylinuxstb: Okay, you're right. I'm actually getting proper error messages now from the NAND driver. Maybe it had something to do with the RAM setting.
00:36:54linuxstbYes, having the RAM set wrongly would do bad things.
00:38:09robin0800jhMikeS: it only says charger present now and a voltage of 4.9v!
00:38:15 Join tessarakt [0] (
00:38:15linuxstbMarcGuay: What's the status of the C100? Is the LCD working?
00:38:21 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
00:38:26jhMikeSrobin0800: Status says?
00:38:36 Join itcheg [0] (i=42b0fb98@gateway/web/ajax/
00:38:44jhMikeSState rather
00:40:52robin0800last powerhist 4.911v battery 100% est remain 706 min
00:40:57MarcGuaylinuxstb: LCD & buttons work. It's erroring out on disk_mount_all().
00:41:54MarcGuay(v1 with the NAND interface)
00:43:03MarcGuayParticular error is "No partition found".
00:43:27 Join domonoky1 [0] (
00:43:38jhMikeSrobin0800: I need to know if you see "State: Error" under the power status page.
00:43:40JdGordonpixelma: i went to bed thinking "why the hell would that patch work" :D
00:43:58pixelmait does
00:44:35pixelmabut since testing out resume I found a bug which I traced back to r19705...
00:44:38JdGordonit shouldnt :p
00:44:50linuxstbMarcGuay: Now you need to hope shotofadds reappears...
00:45:22MarcGuaylinuxstb: :) Been a while... I'll poke at it myself but don't imagine I'll get far.
00:45:51pixelmaJdGordon: interested? Enough coffee yet? ;)
00:45:55robin0800jhMikeS: where is the power status page?
00:46:02linuxstbMarcGuay: I think he's said he's been making good progress with it, but I'm not sure if he's published any of those improvements.
00:46:33JdGordonpixelma: wirte it up... but no.. still asleep :p
00:46:45MarcGuaylinuxstb: Maybe I can see if the test pcm beep code will run..
00:47:20jhMikeSrobin0800: System|Debug (Keep Out!)|View Battery. It has several pages you can switch to (probably using LEFT/RIGHT/UP/DOWN or something).
00:48:53pixelmaJdGordon: you need to have "follow playlist" enabled. Then start playback, enter file browser frome the WPS without stopping playback (happened to me even when entering file browser via the menu), then return to WPS with resume button, then try to stop
00:50:00pixelmastatusbar and a splash flashing by and you're back in the WPS. The entering file browser once step seems to be needed
00:50:12 Quit jgarvey ("Leaving")
00:50:20jhMikeSFrankly I've seen retailos not finish on e200 if the battery was extremely low and it need to be plugged again.
00:50:33robin0800jhMikeS: no error messages and voltage goes from 3.8 to 4.9 in about 10 seconds can't be right!
00:51:28jhMikeSrobin0800: c200 doesn't show voltage accurately when plugged.
00:51:37***Saving seen data "./dancer.seen"
00:51:38pixelmaJdGordon: first seen on c200 (with r19723) and traced back to r19705 in an H100 sim (amiconn confirmed it happening on H100 target too)
00:51:54jhMikeSrobin0800: State: say what though after plugging?
00:52:11JdGordonpixelma: ok, ill have a look after some breaky
00:52:49pixelmathanks, if you want I can file a bug report but didn't know how to summarise nicely so someone reads it
00:54:07 Nick fxb is now known as fxb__ (
00:55:18JdGordona report would be good, incase i dont get around to it today
00:55:48robin0800jhMikeS: rockbox info says its charging battery says no as does wps
00:55:52 Quit jhulst (Remote closed the connection)
00:56:29PaulJamJdGordon: if you find the time, could you also look at FS #9772. might be related to your statusbar rework.
00:56:35jhMikeSrobin0800: I'm not concerned with what the UI says. The "State: " line in the battery info is the official indicator.
00:56:46pixelmaJdGordon: have a suggestion for a title?
00:57:16JdGordonPaulJam: yeah, saw that... hopefully today also
00:57:21JdGordonpixelma: not really
00:57:27robin0800jhMikeS: that says charger present
00:57:29*JdGordon wanders off to find food
00:58:04 Quit ze ("new gpu")
00:58:07jhMikeSrobin0800: Argh. State: not Charger:. it's the line just below it.
00:58:45robin0800jhMikeS: Missing the error!
00:59:27 Quit domonoky (Read error: 110 (Connection timed out))
00:59:43 Join Xerion [0] (
01:01:25 Quit tyfoo ("Carpe diem")
01:01:49 Quit Schmogel (Read error: 104 (Connection reset by peer))
01:03:18jhMikeSrobin0800: If you go to the battery graph, then press up or down (I'm not sure for c200), there are three more pages to it.
01:03:46*jhMikeS double checks the patch for something lunatic he did
01:05:52jhMikeSDoesn't seem to be a problem with #defines. Press whatever is PREV/NEXT for c200 to switch views.
01:08:21 Quit itcheg (" ajax IRC Client")
01:09:19 Join Unhelpful [0] (n=chshrcat@rockbox/developer/Unhelpful)
01:09:34robin0800jhMikeS: yes what do you want to know .....the status line is missing, voltage deltas are <20mv and the third screen says its charged
01:13:24 Quit bmbl ("Woah!")
01:13:51 Quit MarcGuay ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
01:15:10 Quit gevaerts (Read error: 60 (Operation timed out))
01:15:54 Quit tessarakt ("Client exiting")
01:17:20jhMikeSrobin0800: Did you build clean? The view is definitely there on my e200 and "..|| defined(SANSA_C200)" is there
01:18:51robin0800jhMikeS: yes deleted rockbox folder and started fom there
01:19:44pixelmathere seems to be 4 screens in "view battery" on my c200 (don't have the charging patch applied) - first is the one with the graph, pressing "down" gives 3 more
01:20:10robin0800pixelma: correct
01:20:46 Join tvelocity [0] (
01:21:19robin0800pixelma: please try it if you can
01:21:41pixelmajhMikeS: you don't have any info a bit lower on the screen? The c200's display is only 80 pixels high
01:22:13jhMikeSpixelma: how many lines fit? it's only a few lines
01:23:23robin0800pixelma: I have seen it when it first started @ 23:20 but not since then
01:23:55*jhMikeS forgot how high the sysfont is...8 pixels?
01:24:03pixelmathe screens in SVN seem to be made for 8 lines (probably for Archos display, with sysfont) - all should fit
01:24:12pixelmayes 8
01:24:27jhMikeSI only have 5 lines of info
01:24:37Unhelpfulanybody happen to know if the archos sims simulate the non-square pixels?
01:24:48pixelmaso much for that idea then
01:25:52robin0800jhMikeS: I only have three lines of text
01:26:44pixelmaUnhelpful: in general they don't. There is a correction for it which some plugins use (for the screenshots only?), e.g. clock, but that's not "squeezing" the complete display
01:26:53robin0800jhMikeS: the voltage delta has eight lines
01:27:21Unhelpfulok, then i imagine the best i can really do for aspect ratio correction is test that it builds and doesn't act completely insane. :/
01:27:48jhMikeSI just made a c200 binary, and the state strings are present.
01:28:32robin0800jhMikeS: What are they?
01:28:39Unhelpfuli have a feeling that test_greylib_bitmap_scale won't work on archos as-is... it uses a screen-sized greylib overly, i seem to recall amiconn saying that was pushing things pretty close on lowmem targets already
01:29:23pixelmarobin0800: if battery still reports +4.6V when plugged I don't trust it and am not keen to try. I already use the Rockbox USB stack which is not stable yet...
01:29:30 Join goffa [0] (n=goffa@
01:29:51jhMikeSrobin0800: Disables, Error, Charging, Discharging, <unknown>
01:29:58pixelmaJdGordon: FS #9773 for your convenience ;)
01:32:01jhMikeSapps/debug_menu.c should have that screen code starting line 1757
01:32:07pixelmaUnhelpful: I'd be willing to test for you on my Ondio but not... now ;)
01:32:45Unhelpfuli at *least* need to make it allocate the grey buffer and the bitmap dynamically from plugin buffer
01:33:31Unhelpfulfor right now, it just has an array declared to use as the greylib buffer, and that can't help the memory crunch any.
01:34:53robin0800pixelma: 4.6v! mine says 4.9v but idon't believe it but i think that must be what the patch is seeing
01:36:48robin0800jhMikeS: when unplugged 3.85 v and 55%
01:36:57 Join Xerion_ [0] (
01:37:13Unhelpfulthe sim looks squished vertically, if i scale it up vertically by 1.25, the image looks right again
01:37:58 Join MU{lappy} [0] (n=Militant@rockbox/developer/Unhelpful)
01:38:01pixelmawell, I suppose your monitor has square pixels
01:38:46jhMikeSrobin0800: When plugged you aren't seen the actual battery voltage. It's ignored as meaningless (which is is for charging decisions).
01:39:07Unhelpfulalso, the measured dimensions in the screenshot, after scaling it, have the same aspect ratio as the image :)
01:39:46pixelmaUnhelpful: but that sounds correctly for the aspect ratio corrected scaling - if you compare the chessboard with it for example which is also square on the Archos display
01:40:40 Part toffe82
01:40:46Unhelpfulwell, the plugin builds for target with the static greylib buffer, so at the very least, code+overlay fit in the plugin buffer
01:41:33 Quit goffa_ (Read error: 110 (Connection timed out))
01:41:52Unhelpful...actually, it looks like the plugin already uses a static load buffer, as well. if both of those fit when it compiles, it should work :)
01:41:59pixelmaI remember "fire" not working right when using the whole screen so had to drop a few lines from the sides
01:42:51*jhMikeS worked out the code himself and arrived at similar procedures to what bertrik did. There's no essential difference so if the other patches charged, this should too.
01:43:38jhMikeSbut if the build is weird and not showing correct screens, etc, I'm not sure what to make of it.
01:44:01robin0800jhMikeS: so it should always start? how does it decide when to stop?
01:45:29robin0800jhMikeS: I had no trouble with bertrik's patch
01:45:49advcomp2019funman (for the logs): i do not have a 4gb clip.. i only have 1gb and 2gb clips
01:46:52Unhelpfuli know you said you can't *now*, but when you get a chance, would you mind trying and loading a nice color or greyscale bitmap with test_greylib_bitmap_scale?
01:47:18Unhelpfulpretty much anything should work, the bees image i like to test with was obviously wrong in sim until i took a screenshot and scaled it
01:49:12 Quit Unhelpful ("Leaving")
01:50:02 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
01:50:24 Quit parafin (Read error: 60 (Operation timed out))
01:50:26 Quit Xerion_ (" ")
01:51:24 Join itcheg [0] (i=42b0fb98@gateway/web/ajax/
01:53:23 Quit MU{lappy} (Read error: 104 (Connection reset by peer))
01:53:39daurnmy fuze now has rockbox on it
01:53:42daurnanythign need testing etc?
01:53:46 Join parafin [0] (
01:54:40 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
01:54:48 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.5/2008121623]")
01:54:50 Quit Xerion (Read error: 110 (Connection timed out))
01:56:21jhMikeSrobin0800: It starts when plugged and stops when the indicators say to. Same thing.
01:57:08robin0800jhMikeS: what indicators?
01:57:33 Join aurix_lexic1 [0] (
01:58:29jhMikeSrobin0800: CHG_TMPHIGH | CHG_ENDOFCH. There is one twist with checking CHG_status but I don't recall seeing this set on my e200.
01:58:38 Join Unhelpful [0] (n=Militant@rockbox/developer/Unhelpful)
01:59:23 Quit Aurix_Lexico (Read error: 60 (Operation timed out))
01:59:57pixelmaUnhelpful: do I need to still add the plugin to viewers.config, stuff like this?
02:00:32Unhelpfulthe plugin's already in viewers.config, and the patch adds it to SOURCES as well - i'll remove that part before committing.
02:00:38pixelmaI'm really tired though now (2AM here) and even putting a build onto the Ondio takes quite some time
02:01:57Unhelpfulno hurry, i'll ask somebody else if i see anybody wake up, and i'm going to be here for another 11hr or so
02:03:17robin0800jhMikeS: Until one of those is true it shouln't stop charging is that correct?
02:04:29jhMikeSrobin0800: Right, it should keep going until one is set
02:04:39jhMikeSat least one
02:05:25 Quit itcheg (" ajax IRC Client")
02:05:27 Quit Thundercloud (Remote closed the connection)
02:07:36robin0800jhMikeS: no easy way to find if one of those is set?
02:10:26jhMikeSI don't have a display because reading them seems to clear them. Only powermgmt should touch that register.
02:15:12robin0800bertrik has a /* read status to clear initial end-of-charge indication */ in his patch
02:15:36jhMikeSI've got that too
02:16:05robin0800jhMikeS: so its not that then
02:17:07 Join jfc^2 [0] (
02:17:11 Join CryptA [0] (n=417fc129@gateway/web/cgi-irc/
02:17:20jhMikeSIt might be the wait for CHG_STATUS
02:17:51 Join Rob2222 [0] (
02:18:36 Join neddy1 [0] (n=john@nat/sun/x-1ee5b6ef0bfaa5a3)
02:18:44robin0800jhMikeS: its getting late here going to unplug, leave on & start again tomorrow
02:19:55 Join itcheg [0] (i=42b0fb98@gateway/web/ajax/
02:21:00CryptAGreetings, wonder if someone can provide edit access to "CFModGuide" so I can add my 2cents?
02:22:33 Join Bensawesome [0] (n=Bensawso@unaffiliated/bensawsome)
02:22:51 Quit moos ("Rockbox rules the DAP world")
02:23:39 Join goffa_ [0] (n=goffa@
02:28:01linuxstbCryptA: Have you registered on the wiki?
02:28:41linuxstbCryptA: I see you have, but not with the required name. Please re-read the registration page....
02:31:59CryptAlinuxstb: Thanks, duely noted.
02:32:51 Quit Bensawsome (Connection timed out)
02:33:08 Quit robin0800 (Remote closed the connection)
02:33:29 Quit jfc (Read error: 110 (Connection timed out))
02:35:42 Quit Rob2223 (Read error: 110 (Connection timed out))
02:35:55 Quit goffa (Read error: 110 (Connection timed out))
02:36:44 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
02:44:51 Join Xerion [0] (
02:46:14 Join ze [0] (i=ze@
02:46:27 Nick Bensawesome is now known as Bensawsome (n=Bensawso@unaffiliated/bensawsome)
02:48:53 Quit neddy1 (Read error: 110 (Connection timed out))
02:51:27 Quit itcheg (" ajax IRC Client")
02:51:40***Saving seen data "./dancer.seen"
02:52:05 Quit gevaerts (Read error: 113 (No route to host))
02:55:43 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
02:57:10 Join axionix_ [0] (
03:01:05 Join Darksair [0] (n=user@
03:01:20 Quit saratoga ("CGI:IRC (EOF)")
03:03:31 Quit gevaerts ("le")
03:04:54 Quit reacocard (Remote closed the connection)
03:05:17 Quit lucent (Read error: 60 (Operation timed out))
03:08:07 Quit axionix (Read error: 110 (Connection timed out))
03:09:57 Join unstable_ [0] (
03:11:36 Quit domonoky1 (Read error: 54 (Connection reset by peer))
03:13:00 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
03:16:26 Quit tvelocity (Remote closed the connection)
03:16:52 Quit unstable (Read error: 60 (Operation timed out))
03:18:25 Join Strife89 [0] (n=michael@
03:20:17 Join massiveH [0] (
03:22:44 Quit BXCracer (Remote closed the connection)
03:24:09 Quit massiveH (Client Quit)
03:24:16 Quit havien (Read error: 110 (Connection timed out))
03:28:19 Join unstable [0] (
03:29:11 Quit unstable_ (Read error: 113 (No route to host))
03:31:09 Join roebeet_ [0] (
03:33:12 Join MOHDMACH [0] (
03:33:37MOHDMACHHello once again
03:34:45 Join kachna|lappy [0] (
03:35:29 Quit DerDome (Nick collision from services.)
03:35:32 Join DerDome1 [0] (
03:35:42 Nick DerDome1 is now known as DerDome (
03:35:44MOHDMACHI have a .wps file which has parts I am trying to use but I don't understand how it works in the first place. So the lines I highlighted in here: are the ones confusing me.
03:36:02 Quit kachna (Read error: 104 (Connection reset by peer))
03:37:51 Join itcheg [0] (i=4c6eb511@gateway/web/ajax/
03:39:12Unhelpfulwhat exactly is the question? line 12 enables the display of the viewports labeled v, k, s, p based on various conditions. line 39 defines a viewport labeled c to be used at some other point
03:39:42 Join neddy [0] (n=john@nat/sun/x-30215311ac75d8a9)
03:39:54MOHDMACHThe highlighted has the playback mode tag and the volume tag combined and the question is how do they work together and what is the purpose of combining them.
03:40:47MOHDMACHAlso I read the CustomWPS guide and could not figure out what the %?C is for.
03:41:30 Quit itcheg (Client Quit)
03:41:32Unhelpfulthat would be the conditional for the presence of album art for the current track
03:43:09Unhelpfulas far as "how they combine", they don't. after the conditional itself, there's some more text, <a|b>, and it uses either a or b based on whether the condition is true or not. you "combine" the conditions by putting the second one inside the <>, as is done on line 12
03:44:38 Join MOHDMACH_ [0] (
03:44:43MOHDMACH_Viewports aside the images also have the same letters so would that make the .wps file invalid wouldn't it?
03:45:00MOHDMACH_By the way it works.
03:45:41Unhelpfulline 121 is a comment.
03:46:38MOHDMACH_I was referring to 's' but ok.
03:47:11Unhelpfuli was looking at the highlighted lines
03:47:42MOHDMACH_Oh sorry I was actually referring to line 150.
03:47:43Unhelpfuli only see ony xl tag for 's'
03:48:10Unhelpfulyes, there's no other %xl|s| that i can see?
03:48:38MOHDMACH_There is a %Vds line 12.
03:49:02Unhelpfulthat's a display tag, and it's a viewport, not an image.
03:49:43MOHDMACH_Ah ok.
03:50:04MOHDMACH_But what is the purpose of the volume tag being combined with the playmode tag?
03:51:18Unhelpfulto display whatever it displays only while the volume is being changed and music is playing?
03:52:22MOHDMACH_So in other words the volume bar won't appear while the music is paused?
03:52:32 Join goffa [0] (n=goffa@
03:53:21Unhelpfuli don't recall what all the different modes are, but something *like* that.
03:53:21 Quit Strife89 ("Going to bed. :)")
03:53:21 Quit CryptA ("CGI:IRC")
03:54:44MOHDMACH_What is the purpose of >> in line 12.
03:54:48 Join RycPat [0] (n=417fc129@gateway/web/cgi-irc/
03:55:16MOHDMACH_I did a search for it in the CustomWPS and it does not show anything.
03:55:51Unhelpfulof course there's not, that's the > for two different conditional tags. they nest.
03:57:13MOHDMACH_Oh sorry.
03:57:43Unhelpfuleach > matches to a < somewhere, you have to work from the outside in and pair them up
03:57:59MOHDMACH_It looked like something else when I was going through it.
03:58:30 Join itcheg [0] (i=62db4767@gateway/web/ajax/
03:59:31MOHDMACH_Also what is needed for a volume tag to work with conditional viewports other than preloaded bitmaps, conditional viewport tags, and an actual image?
03:59:49 Quit MOHDMACH (Read error: 110 (Connection timed out))
04:01:06 Join toffe82 [0] (
04:01:19RycPatCan someone please provide me with write permission to the wiki? I would like to upload my info to CFModGuide.
04:01:23 Quit PaulJam (".")
04:02:12 Join blkhawk- [0] (
04:03:44 Quit goffa_ (Read error: 110 (Connection timed out))
04:09:21UnhelpfulMOHDMACH_: nothing. the %Vl tag defines a viewport that contains everything on the lines below it, until another viewport tag is reached. you can put whatever you want to inside the %Vl tag, and then display the viewport with the matching %Vd tag.
04:10:47 Join krazykit` [0] (
04:10:59MOHDMACH_Thanks I kind of get what you are saying.
04:12:49MOHDMACH_Except I really can't make sense of what line 12 is saying.
04:14:37Unhelpful%mp contains four sections, and displays one of them based on the current playmode
04:14:37Unhelpfulerm, five, sorry :)
04:15:20Unhelpfulif playing, that line displays %?mv<%Vdv|%Vdk%?C<%Vds|>>, which means contains a further condition for whether or not volume is changing
04:16:10Unhelpfulit looks like that's actually what it displays if the state is anything other than fast-forward or rewind
04:16:34MOHDMACH_Ok thanks.
04:16:52MOHDMACH_I have to go.
04:17:00MOHDMACH_Thank you very much for the help.
04:17:06 Quit blkhawk (Connection timed out)
04:17:10 Nick blkhawk- is now known as blkhawk (
04:17:18 Quit MOHDMACH_ ()
04:19:22 Join Erufu [0] (
04:19:29 Join fdinel [0] (
04:20:17 Quit Bensawsome (Read error: 104 (Connection reset by peer))
04:21:34 Part Erufu
04:24:47 Quit krazykit (Read error: 110 (Connection timed out))
04:26:09 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
04:28:16 Join reacocard [0] (
04:35:08 Quit miepchen^schla (Read error: 110 (Connection timed out))
04:37:51 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
04:38:55 Join Erufu [0] (
04:46:25 Join Barahir_ [0] (
04:47:12UnhelpfulErufu: nobody said you had to *leave* -community, it's just that that's for social chat
04:47:27 Join lucent [0] (
04:47:47ErufuI'm an IRC newbie, and I always seem to be inevitably kicked off of it for some reason or another
04:47:53Erufujust tryin' to not be an ass =P
04:48:14lucentwhat did I walk into
04:48:18Unhelpfullucent: did you ever get any results with that bmp chunk size patch?
04:48:21 Join nbeebo [0] (
04:48:32lucentUnhelpful: yes! it did prevent the bad pixel values on my 8gb fuze v1
04:48:55Unhelpfulit's a very, very weird bug :/
04:49:10lucentUnhelpful: I don't approve or disapprove of the patch, just saying what happened
04:49:53Unhelpfulthe patch only covers up whatever the real bug is, in my opinion. i won't commit it unless forced. ;)
04:50:56lucentdo we have a guess on *why* it covers up the bug?
04:51:25lucentI gotta say that data access on the Fuze with current rockbox firmware is not the most reliable
04:51:40lucentit doesn't appear to work if reading files above 1G
04:51:42***Saving seen data "./dancer.seen"
04:51:48lucentI need to test this again though
04:51:54 Join Willwolfe [0] (
04:56:20Unhelpfulthat i *don't* know. my guess is that if changing read size hides it, it's a fat or storage driver bug, but it's kind of creepy how consistent it is. :/
04:56:57 Part nbeebo ("Ex-Chat")
04:59:18ErufuI don't know what I'm talking about, but it could be a bank-switching bug
04:59:46Erufuseems the best way to use multi-gigabyte addressing on a 32-bit processor
05:00:45 Join blkhawk- [0] (
05:02:37lucentI'm going to try in the dark here, adding delays and make SD external work on 8gb fuze
05:03:37Unhelpfuli don't think we *can* access past the first GB of the internal storage at present... and isn't the fuze internal still via SD?
05:03:45 Quit Barahir (Read error: 110 (Connection timed out))
05:05:06UnhelpfulErufu: the problems that larger reads cause in the bitmap loader look more like it's just missing a byte or two, here and there, with the bytes following it shifted over accordingly. don't ask how *that* can happen, i don't know :/
05:06:08Erufuperhaps it's a word-biasing problem, that is, only being able to access data by words or doublewords, and not by individual bytes
05:06:35Erufuagain, I really have no idea what I'm talking about, just starting to read the prodigious amount of documentation
05:10:37Unhelpfulwell, ARM at least will not silently corrupt data if you ignore its alignment requirements
05:11:21 Quit GodEater_ (" ajax IRC Client")
05:11:51Erufuwell, I was just guessing. If it's larger reads, then perhaps it's the DMA (if it exists on the device) aliasing like that for faster block processing
05:12:21Erufumost of my ARM experience is from the GBA, which is ARM7, so...
05:12:30 Join GodEater_ [0] (i=c2cbc962@rockbox/staff/GodEater)
05:16:36 Quit blkhawk (Read error: 113 (No route to host))
05:16:43 Nick blkhawk- is now known as blkhawk (
05:16:58Unhelpfulit's the larger reads that fail... the "small" reads that succeed are 8 pixels worth of bitmap data at a time, so 24B, assuming RGB24
05:18:05Unhelpfuli don't think anybody's really tried anything in between 8p and the whole line
05:19:01Erufumight be a limitation of the video hardware
05:19:06Unhelpfullucent: do you still have that patched source around? and would you mind setting BM_MAX_WIDTH to something larger, say 16? it needs to be a multiple of 8 in case a mono bitmap is being read
05:20:02lucentUnhelpful: will do
05:21:05Unhelpfuli'd be interested in seeing at what point bitmaps start to corrupt... the icons seem to be ok, it's only the WPS background, the menu background, and the progress bar that seem to be corrupted
05:30:26 Quit RycPat ("CGI:IRC")
05:34:04 Quit roebeet_ ("Ex-Chat")
05:36:21 Part toffe82
05:48:22 Join MOHDMACH [0] (
05:50:11 Quit esthar (Read error: 60 (Operation timed out))
05:53:51 Quit Willwolfe ("ChatZilla 0.9.84 [Firefox 3.0.1/2008070208]")
05:57:24 Quit Horscht ("")
06:00:07 Part MOHDMACH
06:13:55 Quit pixelma (Read error: 110 (Connection timed out))
06:15:14 Quit amiconn (Read error: 110 (Connection timed out))
06:15:42 Quit itcheg (" ajax IRC Client")
06:15:56 Quit aurix_lexic1 ("Leaving.")
06:16:05 Join pixelma [0] (n=pixelma@rockbox/staff/pixelma)
06:16:10 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
06:20:46 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:25:37Erufucuriously, the "recovery mode" instructions listed on the sansa view wiki don't work for my 16GB version
06:28:59advcomp2019Erufu, are you turning off all the way?
06:29:25Erufupretty sure I am
06:29:42Erufuaha, there we go
06:29:48Erufumy fuckup
06:31:20advcomp2019Erufu, i have a view too.. so i know things about it already
06:33:18Erufua good addendum to that wiki would be to mention that the backlight around the selection wheel won't light up completely if you completely turn it off
06:38:45*daurn asks: who works on v2 sansas/fuze port here?
06:49:31lucentI help test
06:49:38lucentkugel contributes a lot
06:49:51lucentso do funman and fdinel
06:51:47***Saving seen data "./dancer.seen"
06:55:01lucentshoot, I don't know where to go to make sd work on the 8gb fuze
06:57:49 Join Lss__ [0] (
06:57:50 Quit Lss (Read error: 131 (Connection reset by peer))
07:01:19lucentooOh I got it
07:02:06 Quit jhulst (Read error: 113 (No route to host))
07:03:51 Join toffe82 [0] (n=chatzill@
07:04:43toffe82What is the use of the option record on a video ipod ?
07:06:44lucentyay I have CSD data for my 8gb card
07:11:10lucentSansa Fuze8gb CSD for ISD: 0x92404018, 0xBEFBCFFF, 0x5F5983D3, 0x00260032
07:11:13lucentSansa Fuze8gb CSD for 8gbClass6SDHC: 0x0A40003E, 0x3C1B7F80, 0x5F590000, 0x400E0032
07:11:18Unhelpfultoffe82: there's at least one microphone attachment that's verified working with 5.5G... that's the same as video, yes?
07:13:09UnhelpfulIpodAccessories says the the "!XtremeMac" Micromemo works... presumably there are others as well
07:14:54toffe82there should be a note for this in the manual, that it is for external micro or line in as there is no use if it is not connected to anything
07:15:15 Join midkay_ [0] (
07:24:01lucentI got it working by messing with the value that is used in mci_delay
07:25:18 Part toffe82
07:28:12 Quit midkay (Read error: 110 (Connection timed out))
07:33:33Unhelpfulwhat, the device data you read?
07:35:45 Join matsl [0] (
07:37:37lucentUnhelpful: there's some delay routines in the SD code for Sansa Fuze
07:38:00Unhelpfuland changing that is what enabled you to read that CSD data?
07:38:10lucentoh the two are kind of unrelated
07:38:26lucentI'm having issues on 8gb fuze that kugel can't reproduce on his 4gb fuze
07:38:49lucentwhen I insert an microSD card the kernel panics
07:39:04lucentkugel suggested I mess with the mci_delay
07:39:24lucentCSD was for someone else who was asking, I think it was funman
07:40:46lucent65535 is the default value for mci_delay in the code already, but it doesn't work for me
07:41:20lucent193750 (a longer delay) does not work either. But, I find that 196608 does work
07:48:14daurnlucent: well, I just got an 8gb, rockbox boots, but nothing works
07:48:39daurncan read directorys, but not files it seems (though maybe this is just over the 1gb boundary I heard about?)
07:48:55daurn(this is with internal storage)
07:49:27daurntried playing doom, it gets to the menu for the wad, but froze/turned off after I moved menu selections (it did play the sound though....)
07:49:42daurndo you know about the dirs ## MUSIC that you can see in rockbox?
07:50:54Unhelpfuldaurn: i'm not closely involved in fuze, but what i've heard a few times is that apparently we can't read past the first 1GB, right now... i think some devs have "fixed" this by creating a smaller filesystem?
07:52:18lucentdaurn: the ##MUSIC## and such directories have the Volume Label bit set in the fat filesystem
07:52:32lucentLinux vfat driver ignores these entries
07:52:47lucentyou can recompile a fuse fat driver to read these
07:54:22 Quit Darksair (Read error: 104 (Connection reset by peer))
07:59:03 Nick Lss__ is now known as Lss (
08:00:03 Quit BHSPitLappy (Remote closed the connection)
08:01:30 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
08:03:51Erufuawesome, my laptop just utterly died
08:03:56_Auron_it hates you
08:04:45Erufuor somethin'
08:05:51Unhelpfulhrm... we don't support mutexes on hwcodec devices? or are they just not in the pluginlib?
08:06:11*lucent goes through the mind numbing task of regression testing
08:06:40lucentGot it!
08:09:22jhMikeSUnhelpful: Only mpegplayer was using them so far.
08:10:24 Quit XavierGr ()
08:11:25Unhelpfulhrm. i've been working on getting pictureflow.c on mono targets, with the eventual goal of supporting archos, and i'm stuck on missing mutex and buffer APIs in plugin_api
08:11:46Unhelpfuli see that buffering.c is not even built on swcodec, so this may not be fixable without seriously reworking PF
08:11:47amiconnbuffer api?
08:12:02amiconnWhy the heck does PF need the buffer api??
08:12:35amiconnThis of course won't work on archos. Hwcodec uses a different playback engine
08:12:51Unhelpfulthat's how it loads covers for display... it's basically abusing bufalloc as a malloc to avoid having to manage its own memory
08:13:07amiconnAs for the mutexes, there someone obviously didn't pay attetntion
08:13:49Unhelpfulso, mutexes will be fine if i just remove the SWCODEC check around them?
08:14:05Erufumalloc on an embedded system? that strikes me as odd...
08:14:19amiconnWhen functions are added to the plugin api, they shouldn't be narrowed down, i.e. they should be added for all targets which provide that function
08:15:07amiconnErufu: There is no malloc in rockbox, but some plugins need to do some kind of malloc-like memory management
08:16:53ErufuI figured malloc was unnecessary, so I was just confused why any allocation system was used
08:17:40UnhelpfulErufu: how else do you manage variable-sized chunks of data that you're loading and unloading at random? that's basically what PF needs. :/
08:18:13lucentI'm pretty excited about making a workaround for this
08:18:14Erufu"at random" in a linearly-executed environment
08:18:15Unhelpfuli could make them fixed-size if we don't want to support disabling scaling - could just have an array of chunks big enough to hold a max-size cover.
08:18:31ErufuI figured you'd just use a junk box, or sandbox of sorts, for temporary data
08:18:32lucentwhat's our preference, calling mci_delay a few times or making a new delay function inline?
08:19:02UnhelpfulErufu: plugins sometimes work a little differently, or are ported from other codebases
08:19:42Unhelpfuland yes, "at random" in this case, as the user can scroll around in a big list of cover art, and X covers to either side of the current one need to be loaded
08:20:24Erufuwell, I would see that as three necessary data segments, and loading and unloading into those three established segments as necessary
08:20:33Erufuwithout the overhead of a memory management system
08:20:54Unhelpfulexcept that they can be of arbitrary sizes, and there are more than three
08:21:14Unhelpfulyet another reason to at *least* force AA scaling to keep below a max size, i think.
08:21:18 Join kugel [0] (n=chatzill@unaffiliated/kugel)
08:22:05Erufuwell, either way there can only be a static x number of cover art on the screen
08:22:19Erufuseems pretty straight forward to me
08:22:40Unhelpfulum, is get_metadata truly unavailable on swcodec as well, or is that another quirk of the plugin API?
08:23:58Unhelpfulthe issue is that, at present, these cover arts can be of arbitrary dimensions. i think the solution is to enforce a maximum size, and scale down anything above that on load, whether the user has resizing enabled or not.
08:24:26Unhelpfulpretty much anything else leaves us back at abusing bufalloc, or having to implement some kind of real memory management for the plugin buffer
08:25:10 Quit z35 (Read error: 110 (Connection timed out))
08:26:24lucent8GB Fuze V1 microSD external panic fix:
08:26:29Erufuonly one plugin can operate at once, right?
08:26:44kugelUnhelpful: it's available, and used in pictueflow
08:26:49lucenthi kugel
08:27:11lucentsee my pastebin there, I did like you said and increased the delay to fix the panic on microSD insert
08:27:26kugelwait a second
08:27:28 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.5/2008121622]")
08:27:37lucentmaybe you can make some sense about why I need 3x delay there
08:28:20 Join ender` [0] (
08:28:27amiconnUnhelpful: Well, normally the slides should be (roughly) the same size, and iiuc there's a slide cache with a fixed number of entries
08:28:48amiconnSo this could just put the bitmap data into the array, and be done
08:28:57 Quit BigBambi (Read error: 113 (No route to host))
08:29:26Unhelpfulamiconn: that's true, but they are *allowed* to be of arbitrary size, at present. i think i need to force scaling on for "oversized" slides, pretty much
08:29:46Unhelpfulalso, it appears that get_metadata was another one i just had to move outside of a #ifdef :)
08:30:01 Join Rob2223 [0] (
08:30:24amiconnIf we don't use a fixed array but allocate it from either plugin_get_buffer() or plugin_get_audio_buffer() depending on whether the music is playing, this should allow running PF while music is playing (on large-mem targets only, of course)
08:30:26Unhelpfuli notice there's no align_buffer on swcodec - do i need to worry about alignment on SH?
08:31:24 Quit shadearg (Read error: 60 (Operation timed out))
08:31:30amiconnYes. SH has the same alignment restrictions as arm
08:31:57 Join kugel [0] (n=chatzill@unaffiliated/kugel)
08:32:00daurnkugel: hey, got ma fuze :D
08:32:44Unhelpfulis there a reason that align_buffer doesn't exist on it, then?
08:32:48 Quit kugel (Client Quit)
08:32:50lucentdaurn do you have a microSD Card?
08:33:11daurnyeah, but depends what you want me to use it for
08:33:16amiconnUnhelpful: The buffering api only exists on swcodec, as already mentioned. Hwcodec uses a different playback engine
08:33:16daurnand what size you want
08:33:17 Join shadearg [0] (
08:33:50amiconnHmm, wait
08:34:17Unhelpfulyes, but align_buffer seems to be of more general use. it's just for aligning the start and end of a region of memory, given a pointer and a size in bytes.
08:35:24lucentdaurn try using one with your Fuze 8gb and see if it works
08:36:14daurnhmmm, rockbox won't boot
08:36:18 Quit Rob2222 (Read error: 60 (Operation timed out))
08:36:20daurnsince I ran doom
08:37:10daurnsome state file or smth I need to delete?
08:37:38 Join kugel [0] (n=chatzill@unaffiliated/kugel)
08:37:57Erufutrying to set up a build environment in kubuntu... I get an error at the end: "C compiler cannot create executables"
08:38:00Erufuany ideas?
08:38:14kugeldid you run with sudo?
08:38:26Erufubut of course
08:38:27kugeli.e. sudo ./
08:38:29scorcheErufu: i assume by "at the end" you mean at the end of running ?
08:38:44Erufu(at the end of
08:39:07kugeldid you also "sudo apt-get install build-essential"?
08:39:08lucentdaurn: at your own risk use mkfs.vfat -F 32 aand specify a sizeto make a 800mb fat partition so you don't worry about stuff... but you know usual warnings it destroys data
08:39:29lucentdaurn: I must quickly correct myself
08:39:33ErufuI didn't see anything telling me to install "build-essential"
08:39:38Erufubut I'll do that now...
08:39:48lucentdaurn: do NOT mess with partitions, just use what's there and make a smaller filesystem
08:40:20*daurn only has one microsd card, (1gb) and its in use, I can take it out temporarily to test something, but can't delete/repartition it etc
08:40:33 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
08:40:34BHSPitLappydaurn, why're you always changing your nick around? You aren't being sneaky, you know
08:40:39lucentonly need to know if panic on insert
08:40:40scorcheit is a bit hard to make executables and such without build-essential ;)
08:40:40Unhelpfulkugel: it's there, but for some reason left out of the plugin API on swcodec targets - probably because of no plugins using it on those.
08:40:56daurnBHSPitLappy: haha, go way >.<
08:41:01Erufugcc doesn't seem to have trouble building anything else, though
08:41:41kugelUnhelpful: picutureflow uses it !?
08:42:04Unhelpfulkugel: there's never been a pictureflow on any swcodec target before
08:42:23kugelUnhelpful: it's there since a few weeks, I introduced it in a patch
08:42:28Unhelpfula couple of weeks ago, there wasn't even one on greyscale targets
08:42:43kugels/swcodec/hwcodec/ ?
08:43:14lucentkugel: does that extra delay break your player?
08:43:27kugelUnhelpful: "there's never been a pictureflow on any swcodec target before" that's wrong
08:43:31lucentif not, maybe, it is a workaround, or is there a deeper problem to fix?
08:44:05*amiconn didn't even know there's a function like this
08:44:13Unhelpful...oh. i see, you're right about that. ;)
08:44:16kugellucent: some times hardware needs some time, and software needs to wait, we have to accept
08:44:24daurnuh, my fuze doesn't show up on my computer any more!
08:44:27Unhelpfulyes, i did mean hwcodec. i need sleep, or coffee, and i'm out of the latter.
08:44:30amiconnI consider buffer alignment trivial enough to not warrant a dedicated function.
08:44:32daurnwtf happended when I ran doom?
08:44:38kugellucent: but I suppose doing it in a loop, with max_tries would be better/safer
08:44:59Unhelpfulamiconn: then i'd better do *something* before pixelma gets up and tries that patch, since it's skipping buffer alignment on hwcodec :/
08:45:17daurndmesg seems to report: sdh: unknown partition table
08:45:28 Join easonand [0] (
08:45:33*lucent groans
08:45:45lucentdaurn please relax
08:45:54*daurn breathes
08:45:55lucentpartitioning WILL BREAK YOUR FUZE
08:45:56lucentdon't do it
08:46:08lucenttreat it like a floppy disk
08:46:30 Nick Barahir_ is now known as Barahir (
08:46:45Erufudaurn: find a really powerful magnet and hope for the best
08:47:21Erufuactually, that probably wouldn't help since you need at least an empty partition...
08:47:31amiconnNo core module seems to ever use that function, just plugins
08:47:43amiconnSo if we really want a function, it should go to the pluginlib
08:48:22Unhelpfuli can see the value in getting it right, once, but shouldn't this be a macro or inline?
08:48:29amiconnOh, and if I'm not totally mislead it seems that mpegplayer does something really nasty
08:48:40 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
08:48:42 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
08:48:58amiconnIt uses CACHEALIGN_BUFFER(), which is a macro defined in system.h that calls align_buffer()
08:49:12 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
08:49:21amiconnThis means it's called circumventing the plugin api (again, if I'm not missing something)
08:49:38 Join kugel [0] (n=chatzill@unaffiliated/kugel)
08:49:51Unhelpfuldoes mpegplayer use ALIGN_BUFFER_WRAPPER, maybe?
08:50:42Unhelpfulanother semiugly hack that can go away with FS #9770 :)
08:50:49amiconnapps\plugins\mpegplayer\disk_buf.c862 disk_buf.size = CACHEALIGN_BUFFER(&disk_buf.start, disk_buf.size);
08:51:42 Quit GodEater_ (" ajax IRC Client")
08:51:45amiconnOops, that was meant to be edited a little before sending, but I think it's comprehensible anyway. It's a c&p from grepping for CACHEALIGN_BUFFER
08:51:51***Saving seen data "./dancer.seen"
08:52:03amiconnThere are only two more hits, both in system.h
08:52:15Unhelpfulmpegplayer.c uses ALIGN_BUFFER_WRAPPER(rb)
08:52:20Unhelpfulit'd fail to link, otherwise
08:54:09 Join GodEater_ [0] (i=c2cbc962@rockbox/staff/GodEater)
08:54:10amiconnHmm, that solves the mystery.
08:54:49amiconnToo much macrofying is obviously hiding things in a bad way :\
08:55:03 Quit shadearg (Read error: 60 (Operation timed out))
08:55:33 Quit kugel (Remote closed the connection)
08:55:34daurnfigured it out
08:55:40 Join kugel [0] (n=chatzill@
08:55:40daurnwhen rockbox trys to boot
08:55:49daurnit says "SD 1 Data Timeout"
08:55:49 Quit Hadaka (Read error: 60 (Operation timed out))
08:55:56daurn(this is without an sd card in)
08:56:00daurnwhats wrong?
08:56:12 Quit avacore (Read error: 60 (Operation timed out))
08:56:22*scorche notes to daurn about the existence of punctuation marks on a keyboard
08:56:36daurnscorche: notice
08:56:40daurnI used some
08:56:58 Join avacore [0] (
08:56:58scorchein other words, stop using the enter key as a substitude for them...
08:57:08lucentplease, less vertical and more horizontal
08:57:10scorchesubstitute too
08:57:32 Quit kugel (Remote closed the connection)
08:57:39 Join kugel_ [0] (n=chatzill@
08:57:44 Nick kugel_ is now known as kugel (n=chatzill@
08:58:08daurnanyway, the question is, why do I get an timeout now, and not previously?
08:59:03lucentif we knew, you wouldn't have that problem.
09:00:42B4gderdaurn: my guess is that there's a bug lurking in there!
09:02:36daurnlucent: can you give me your output of fdisk for your fuze?
09:02:53Zagorbugs? in rockbox?
09:03:08Zagordon't we have a policy against that?
09:03:15B4gderon a fuze even, known to be rock solid
09:04:20Unhelpfulamiconn: there are three functions in firmware/general.c, one of which is align_buffer. any clue at all what the others do?
09:04:50lucentdaurn: god dammit daurn pay attention, there is no partitioning scheme on the fuze
09:05:10lucentit's a block device with a file system
09:05:22scorchelucent: how many times do we have to tell you about the guidelines?
09:06:01*daurn was wondering why the fuze use to automount, but not anymore, and I checked dmesg, the only error was that it has an invalid partition table
09:07:20B4gderdo you ams sansa guys add any partitions for rockbox fun or do you do it superfully ?
09:07:29B4gdersuperfloppy even
09:07:43lucenthas to be superfloppy, Bager
09:07:53B4gder"has to" ?
09:07:54 Join petur [50] (n=petur@rockbox/developer/petur)
09:08:10lucentwe rely on the OF and the OF wants superfloppy
09:08:12ZagorI guess the OF might barf on partitions. and we need OF for usb.
09:08:26B4gderright, it needs to boot properly for usb...
09:09:45kugellucent: did you try a loop
09:10:06lucentkugel: I'm trying to pinpoint which command to do that on, yes I am working on it :)
09:14:09lucentthere's three in the loop
09:14:09 Quit DataGhost (Nick collision from services.)
09:14:09 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
09:14:09lucentline 272 firmware/target/arm/as3525/ata_sd_as3525.c about
09:14:09lucentI don't know how to approach these things
09:14:09Unhelpfulpixelma: whenever you get up, i need to add some buffer alignment to that patch. if you're still itching to test something, it would be a help for somebody with an archos to try chessbox with FS #9770 :)
09:14:09 Quit suom1 (
09:14:09 Quit balou (
09:14:09 Quit lostlogic (
09:14:09 Quit liiwi (
09:14:09 Quit denes (
09:14:09 Quit shodanX (
09:14:09 Quit preglow (
09:14:09 Quit DaCapn (
09:14:09 Quit tim__b (
09:14:09 Quit Kopfgeldjaeger (
09:14:09 Quit Tuplanolla (
09:14:09 Quit Dieterbe (
09:14:09 Join shadearg [0] (
09:14:09kugelI thought CSD was missfunctioning?
09:14:09NJoindenes [0] (
09:14:09NJoinsuom1 [0] (
09:14:09NJoinDieterbe [0] (
09:14:09NJoinKopfgeldjaeger [0] (
09:14:09NJoinbalou [0] (
09:14:09NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
09:14:09NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
09:14:09NJoinliiwi [0] (
09:14:09NJoinpreglow [0] (
09:14:09NJoinDaCapn [0] (
09:14:09NJoinshodanX [0] (
09:14:09NJointim__b [0] (
09:14:27kugellucent: wasn't the return -6?
09:14:58lucentkugel: no, it operates correctly if I put extra delay in the do ... while(!(card_info[drive].ocr & (1<<31)) && max_tries−−); loop
09:14:59 Join Bensawesome [0] (n=Bensawso@unaffiliated/bensawsome)
09:15:24kugelI mean without the delay
09:15:33lucentoh yeah
09:15:37lucentCID would fail
09:15:53lucentif I just loop over CID forever, it never works
09:16:17kugelso, you either put the loop around getting CID, or around what gets the args that getting cid needs
09:16:47lucentoh... I understand better now
09:17:22lucentwhy is it do { mci_delay();
09:17:36lucentshould mci_delay() be the first instruction?
09:18:14pixelmaUnhelpful: good timing and thanks for the warning :) that patch looks dangerous... there's one plugin (for show) that I can't test on Ondio that is also loaded with the overlay loader and that is Rockboy...
09:18:54pixelmaif I remember correctly
09:20:17kugellucent: why not?
09:20:43Unhelpfulwell, the main thing to test with it is whether i've correctly duplicated the changes to the core plugin loader in the overlay loader at all. if chessbox works, rockboy and zxbox should as well.
09:21:02kugelcould probably the last one too, but it works, so let's not worry about it (we can optimize when the port is stable)
09:21:30Unhelpfuleverything i've *tried* has worked on e200, but i've not tried some of the larger plugins, like mpegplayer
09:21:53lucentkugel: my point is that maybe that loop causes the problem
09:22:17kugellucent: the loop I proposed, or the one that's already there?
09:22:26lucentkugel: what's already there
09:22:38kugelwhy'd that cause problems?
09:22:52lucentI think it messes things up because it is not done completing, and then CID cannot ever work
09:23:02lucentI don't know but I am guessing
09:23:55lucentI'm confused why we design parts of code to wait for whatever was behind them to finish
09:24:04lucentinstead of code that runs and finishes before execution to the next part
09:26:10kugellucent: it's getting data from the sd controlloer
09:26:23lucentkugel: I don't think we're waiting long enough for acmd41 to complete
09:26:38kugeland as I already said, sometimes software "is faster" than hardware, then the software needs to wait/retry
09:27:20kugellucent: possibly, but that'd also mean the return val of send_cmd isn't reliable
09:28:04 Join moos [0] (i=Mustapha@rockbox/staff/moos)
09:28:41kugellucent: maybe put the check for "status & MCI_CMD_SENT" before retrieving the response in sent_cmd?
09:28:41 Quit Bensawsome (Read error: 110 (Connection timed out))
09:29:11kugelor, no
09:29:41kugel"else if(status & (MCI_CMD_CRC_FAIL /* FIXME? */ | MCI_CMD_RESP_END))" looks weird to me
09:30:15lucentah, looks alien to me
09:30:24kugellucent: we should probably talk with funman about your findings. send_cmd is essential
09:30:59*kugel wonders why MCI_CMD_CRC_FAIL would indicate a successful response
09:31:00lucentokay so the guess is that send_cmd is returning prematurely?
09:31:36kugellucent: apparently
09:33:02kugellucent: it would surely be helpful which of the send_cmd responses differ with your delay
09:33:18kugel+to know
09:34:03lucenthm, I wonder how to debug this reliably
09:35:08kugellucent: just panicf the response
09:37:19 Quit shodanX (Remote closed the connection)
09:37:23kugellucent: putting the delays before retrieving cid doesn't work?
09:39:26lucentkugel: confirmed, putting delays before retrieving CID results in panic on card insert
09:41:15kugellucent: what if you put them after sending app_cmd?
09:44:21lucentkugel: if 3 mci_delay()'s after sending app_cmd, then NO panic
09:45:01lucentkugel: but then I get a panic after that
09:45:03 Join n1s [0] (n=nils@rockbox/developer/n1s)
09:45:11 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
09:45:13lucent"*PANIC* wait for state TRAN failed"
09:47:11B4gderI guess if I made a system that detected red builds it would be a good start
09:47:21B4gderto trigger actions based on that I mean
09:47:25 Quit easonand ("Leaving")
09:47:27B4gder(switching from another channel to here)
09:48:48Zagoryeah, I'd like that
09:48:57 Join Thundercloud [0] (
09:49:43B4gderperhaps getting the committer names of it as well
09:49:47 Join bobsalad [0] (
09:50:06 Join DrRabbit [0] (n=tuna@unaffiliated/t00na) provides good commit data
09:50:19J-23why you'll not just put their bot here?
09:50:26Zagoruh, we have the commit data
09:51:11B4gderhow on earth would that thing know when our build turns red?
09:51:16Zagorcommits aren't as interesting as the build result.
09:52:34 Quit denes ("KVIrc 3.4.0 Virgo")
09:54:56peturinstant name and shame on a red build?
09:55:08 Nick JdGordon is now known as JdGordon|afk (n=jonno@rockbox/developer/JdGordon)
09:55:53*GodEater_ thinks until we have a working name and shame, we should just always blame petur ;)
09:55:58B4gderyay yay!
09:56:28peturplease do, I'm not committing these days anyway :/
09:56:32*n1s wants a way to see broken manual builds too
09:56:55*petur sees if he can do a commit that breaks manuals
09:58:02lucentkugel: Now I'm getting that panic more often with my patch
09:58:08lucentsomething weird is going on
09:58:15kugelB4gder: a mail to the -dev list maybe?
09:58:38Unhelpfulhrm, i note that the mpeg player plugin seems to hang with FS #9770 :/
09:58:45B4gderI'll provide "the trigger" than we all can go wild doing things with it
09:58:46kugelincluding the appropriate copy of the -cvs mail
09:58:50lucentkugel: why is the value of 0xffff chosen for delay?
09:59:01kugellucent: I don't know
10:00:43lucentkugel: sorry, I don't know what is happening
10:02:30 Join shodanX [0] (
10:04:39lucentI guess the timeout for TRAN is too short
10:05:52kugellucent: just play around with delays/timeouts etc, the source is yours ;)
10:06:27lucentsomething broke when I did svn up
10:07:34lucentooh my patch got applied, yay! about
10:08:10Unhelpfulyay :)
10:13:41Unhelpfule200 shouldn't take a few minutes to start display the menu in mpeg player, should it? :/
10:13:42 Join troy__ [0] (n=toppy@
10:14:29 Quit Erufu ("ChatZilla 0.9.84 [Firefox]")
10:15:21pixelmaa few minutes sound a tad bit too much
10:16:03Unhelpfulit's the only thing i've tested so far that 9770 breaks, then :/
10:16:53 Quit troy_ (Read error: 104 (Connection reset by peer))
10:18:25Unhelpfulworse, it works on sim D:
10:19:57Unhelpfuli'm somewhat more hesitant to suggest you test it on archos now, although chessbox did work for me on e200
10:20:54pixelmathere's no mpegplayer on Archos ;) but I'm currently doing "clean" builds anyway, wanted to test later
10:21:06 Join robin0800 [0] (
10:22:10robin0800jhMikeS: thanks for new patch testing now
10:25:19Zagoronline tts anyone?
10:29:44 Quit Thundercloud (Remote closed the connection)
10:30:07 Nick fxb__ is now known as fxb (
10:30:38 Quit kachna|lappy (Read error: 113 (No route to host))
10:32:17Unhelpfulpixelma: i know there's not... i meant, now that i found something that i've actually broken. i was hopeful that despite how large and invasive the patch was, i might've actually got all of it right.
10:33:19Unhelpfulnow i have to figure out how best to debug something that only fails on target... or, i can just poke around in the source and see if it's obvious.
10:33:26 Join martin__ [0] (
10:38:46pixelmaI'll wait a bit with testing then
10:40:20Unhelpfulwell, the only thing wrong with the scaling-on-mono work is that the scaler relies on align_buffer... and simply doesn't use it on hwcodec. :/
10:43:19*daurn is back
10:43:31daurnanyone able to help me get rockbox to boot on my fuze? it use to work
10:44:37 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
10:45:05mcuelenaerelinuxstb: are you here?
10:45:34 Quit daurn (Remote closed the connection)
10:45:51Unhelpfulwhat do things use to align pointers on hwcodec targets, anyway? is it just done "by hand" in each location?
10:50:17 Quit BHSPitLappy ("Ex-Chat")
10:51:19 Join daurnimas [0] (
10:51:55 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
10:51:56***Saving seen data "./dancer.seen"
10:57:29linuxstbmcuelenaere: Yes.
10:57:52mcuelenaereok, I think I've read most of comments about the licensing issues
10:58:16mcuelenaerewas there any kind of solution for the mtp_dll problem?
10:58:53daurnwhen I try and boot, I it just resets, only for a split second, I can make out something along the lines of: *PANIC* SD : Data Timeout
10:58:58daurnit could be : or 1
11:00:52linuxstbmcuelenaere: It depends what you mean by "solution". I don't think we can make sendfirm GPL'd, which means it can never be incorporated into rbutil. But I'm now slightly more worried about the other code used in MTP_DLL - there's no info at all for it (as far as I can see), and without any license, we can't use it in any way.
11:01:42mcuelenaereso it probably comes down to removing MTP_DLL, rewriting that from scratch?
11:01:47 Quit ze ("fan & kernel")
11:03:09Zagorlinuxstb: I posted a question a couple of hours ago to the irc channel mentioned at but no reponse
11:03:28 Quit daurnimator (No route to host)
11:04:46linuxstbmcuelenaere: Maybe. That still won't solve the GPL incompatibility, but at least the rest of the code will be clean. Maybe the MS SDK has something that can be used as a basis.
11:05:18mcuelenaereI'm looking into Songbird right now, they claim to have a Windows MTP feature..
11:05:56Zagordon't they play drm files too? i.e. they use the binary codecs and thus probably wmp10
11:09:22 Quit jhMikeS (Nick collision from services.)
11:09:28 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
11:09:58Unhelpfulpixelma: when you get a chance, i've revised that scaling patch. the ALIGN_BUF macro feels a bit hackish to me, but it should suffice as a replacement for align_buffer.
11:12:12 Part DrRabbit
11:14:54 Nick daurnimas is now known as daurnimator (
11:18:30linuxstbmcuelenaere: BTW, is sendfirm also used for other ports, apart from the Gigabeat S?
11:18:48mcuelenaereyes, for the Creative ZVM
11:18:56mcuelenaereI mean, it could be used for it
11:19:19mcuelenaerelinuxstb: do you mean the Linux or the Windows version?
11:19:43linuxstbMy last question related to both - I assumed they had identical functionality.
11:19:54 Join Jahgringo [0] (
11:20:20 Quit kugel (Remote closed the connection)
11:21:05mcuelenaereyes, they strive to have the same functionality
11:21:57linuxstb"strive to" ? Does that mean they don't?
11:22:43mcuelenaereI meant they have the same functionality, apart from the implementation
11:22:48mcuelenaerelibmtp vs win32 MTP
11:23:00mcuelenaereso they aren't really identical
11:23:37mcuelenaereand it seems like the Songbird MTP source code has vanished
11:23:50mcuelenaereonly some binaries are still left
11:23:58mcuelenaereor at least that's what I can find
11:25:33 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
11:26:25JahgringoI have a question about rockbox working on different ipods
11:26:28bluebrotherJdGordon|afk: r17315 broke the up button in the quickscreen −− it does now the same as the down button which is worse than doing nothing imo.
11:26:39JahgringoWill rockbox work on ipod classic (6g)?
11:26:46bluebrotherwas this changed intentionally? I just put up a quick patch to fix this and it works fine again ...
11:26:49bluebrotherJahgringo: no
11:26:55JahgringoThank you :)
11:27:02 Quit Jahgringo ("Leaving")
11:27:08scorcheJahgringo: the front page lists the devices rockbox works o....bah
11:27:17 Join Naked [0] (
11:27:30 Quit Naked (Client Quit)
11:27:42pixelmait always surprises me how people find this channel but not *that* info on the front page
11:28:35bluebrothersame for me. Seems the front page isn't readable at all, no matter how you put the information.
11:28:38linuxstbI suppose it isn't that many people, just one every few days... So I'm assuming lots of people do see it on the front page.
11:31:06 Join Naked [0] (
11:31:09 Quit Naked (Client Quit)
11:31:49 Join Naked [0] (
11:32:38 Nick martin__ is now known as igwinnimma (
11:32:46 Nick Naked is now known as Hadaka (
11:34:13mcuelenaereZagor: any luck with #creative?
11:36:03*bluebrother wonders if he should just commit the quickscreen change
11:42:13Unhelpfulit's the fastest way to get comments on it :)
11:42:39bluebrothertrue :)
11:43:30bluebrotherbut it's only like 3 lines changed so I don't expect much problems. And it restores an old behaviour −− no idea why nobody complained for a couple of months.
11:43:31mcuelenaerelinuxstb: does the linux sendfirm also needs to be rewritten (as it's based on a LGPL file)?
11:43:43bluebrothermaybe those h100 aren't used anymore?
11:47:55linuxstbmcuelenaere: Yes, probably. Although that seems trivial, and if I do what I'm planning (merge the two sendfirms, and add extra functionality, such as building dual-boot firmwares), then it will probably end up being unrecognisable compared to the original code anyway.
11:48:44linuxstbI'm happy to take care of the Linux sendfirm, but I'm not equipped for Windows.
11:48:52 Join ze [0] (i=ze@
11:49:23mcuelenaerethen the problem still remains the licensing of the Windows version
11:49:42mcuelenaerewith a bit of luck the Creative source code can get cleared up, but there's still the Microsft SDK then
11:50:20linuxstbI don't think it's solvable - I'm assuimg MS's MTP implementation is the only one for Windows (libMTP seems unusable due to Windows needing usb drivers for each device)
11:51:06linuxstbWe can still keep sendfirm, but just not GPL'd. BSD seems reasonable.
11:51:31Unhelpfulisn't there a libusb for windows?
11:51:39mcuelenaereUnhelpful: yes there is
11:51:53bluebrotheryes, but IIRC you need to install a driver for each device you want to filter
11:52:14mcuelenaerethe Jz4740 utils use that approach
11:52:16bluebrotherthe same problem appears for e200r installation
11:52:49Unhelpfulew. and i assume that's manual, or close to it? :/
11:52:57mcuelenaerebluebrother: what's the problem with e200r?
11:53:47bluebrotherIIRC there was a way to semi-automatic install drivers (well, pre-install them so windows uses them when needed). That would require a windows installer for the tool though ...
11:53:56bluebrothermcuelenaere: you need to install a driver :)
11:54:51*mcuelenaere looks up how to install the e200r
11:55:23 Join Darksair [0] (n=user@
11:57:28linuxstbUnhelpful: The problem is that it also interferes with normal operation of the device under Windows...
11:57:41linuxstbSo you need to install the libusb driver, do Rockbox things, then uninstall it...
11:58:14Unhelpfulthat's very, very sad.
11:59:46mcuelenaerevista seems to have somthing called WinUSB
12:01:44bluebrothersounds nice ... too bad it's not available for xp / 2k
12:02:03mcuelenaereand it probably also requires linking to the SDK..
12:02:42bluebrothertrue, but I assume that at some point this might be available in MinGW's w32api
12:02:55bluebrotheror hope :)
12:07:18Unhelpfulthere's no way to use windows' MTP services via mingw's w32api? :/
12:07:37mcuelenaerenot that I know of..
12:07:59Unhelpfulhrm, i wonder if the shell extensions that let you browse the device could be used to copy a firmware... probably not, i'm not sure how you'd mark it as a firmware file.
12:08:02*bluebrother did a commit and awaits complaints :P
12:09:35 Join casainho [0] (
12:09:52mcuelenaereUnhelpful: RE the app and look up what API it calls and then pass it the FIRMWARE_FILE flag :)
12:09:56casainhohello :-)
12:10:17bluebrotherhellohello :)
12:10:46casainhoI would like to know that if lcd_framebuffer 16 bit colour RGB565, if the most significative bit is red?
12:11:47casainhobecause I have a 12bit LCD and I am trying to convert it from 16 to 12 :-)
12:12:11B4gderah you have it stored 565 in memory?
12:13:04casainhoB4gder: well, I think lcd_framebuffer[y][x] is 565 and I am converting it to 444 before send to LCD...
12:13:41casainhowell, I will not convert lcd_framebuffer[y][x]!! just a copy of it :-)
12:13:42B4gderI get it. It might be clever to avoid introducing a new format, but it's also a minor performance loss
12:14:12Unhelpfulmcuelenaere: no thanks. i'll stay in my bitmap corner. and pictureflow. and plugin_api. this corner thing isn't working.
12:14:32casainhoB4gder: so, do you know if most significant bit of 565 is RED?
12:14:35B4gdercasainho: I don't recall exactly but why not check the macros/code for how the RGB is stored? I think the R is the most significant
12:14:35linuxstbcasainho: Yes, the framebuffer is an array of "unsigned short", the most significant 6 bits are red, then 5 green bits, then 6 blue bits. But your LCD controller doesn't have a 565 mode?
12:15:08B4gderlinuxstb: 17 bits? ;-)
12:15:20casainholinuxstb: okok, thanks :-) −− no, my LCD is the cheapest one and most availably in online shops :-)
12:15:21UnhelpfulB4gder: i'm pretty sure it is, if you're not using the RGB565_SWAPPED format, or however we spell it
12:15:27linuxstbB4gder: Just testing...
12:15:41linuxstbcasainho: Swap "6" and "5" in my sentence...
12:16:04casainholinuxstb: okok, I was thinkg on it :-)
12:16:11casainhookok - I go to do it now ;-)
12:16:34linuxstbcasainho: I would have chosen a 16-bit capable LCD controller though...
12:16:41B4gderme too
12:16:46B4gder444 looks like crap
12:17:18casainholinuxstb: well, I choosed the one I could buy and I also must think one other developers...
12:17:44casainhothat's funny because none of you are buying that hardware
12:17:59linuxstbThat's because we don't like it...
12:18:28bluebrotherif buying a more expensive hardware saves you hours of work it might be worth the money ...
12:18:31casainhoand this prototype is kind of temporary, not need to spend money on it!!
12:19:16linuxstbcasainho: All I mean is that you are "competing" with the commercially sold players Rockbox already runs on (or is portable to).
12:19:33casainhowe are right now in consersations with a big online shop to define a hardware, that shop will be able to sell it worldwide
12:20:05*bluebrother wonders if this is olimex
12:20:05casainholinuxstb: a prototype typically is an ugly thing
12:20:09Unhelpfulanybody with an idea for how to figure out where a plugin is hanging, given that it's only happening on target? preferably something smarter than a million splash(HZ,"i got to here!") calls?
12:20:20casainhobluebrother: yes, is Olimex
12:20:40Unhelpful...isn't here some kind of logf-over-usb thing?
12:20:42linuxstbUnhelpful: logf works over usb on some targets (I've never used it though)
12:20:51linuxstbUnhelpful: Yes ;)
12:21:17bluebrothercasainho: ugly doesn't imply it has crappy hardware
12:21:19Unhelpfulok, so a million logf("i got to here!") calls :)
12:21:47kadobanthat's at least better in that you don't have to wait a (million -1) * HZ cycles i guess
12:21:59ZagorUnhelpful: surely a gdb stub fits in your corner? ;)
12:22:06casainhoguys at Olimex have knowledge to make a good and portable hardware, and there is a lot of online shops, "distribution channels" that seels Olimex hardware :-)
12:23:19casainhobluebrother: yes, right. The prototype should be with the final hardware but this one is not like that, is more to prove that is possible and if we need profs that Rockbox can run on others LCD with better colours, than we can look at Ipods, etc
12:23:28Unhelpfulit's mpegplayer, and it's hanging before presenting its menu, so i somehow doubt i'd be waiting through that many splashes :)
12:23:54Zagorkadoban: splash doesn't have to wait
12:23:58kadobanahh, that's probably easiest then. last time i tried to get logf over usb to work, it ddin't. (although i didn't try too hard 'cause i found my bug soon after)
12:24:35kadobanZagor: ah, i suppose not. good point
12:24:41casainhobluebrother: for me is more a question of having cheap and available hardware, to easily attract new developers
12:24:47bluebrothercasainho: Rockbox runs with LCDs with less quality (monochrome, 2bit like h100 etc). Why is there a need to prove it can get adapted to other LCDs?
12:25:04UnhelpfulZagor: excellent point, in fact, assuming it's not hanging in the middle of drawing something...
12:25:26bluebrotherbut it's your decision. I just think it creates additional work
12:25:28linuxstbUnhelpful: mpegplayer could be trickier, as it's multi-threaded...
12:25:45 Join qwm [0] (
12:25:58casainhobluebrother: sorry, prof that we can make our own hardware, so, easily we can then port Rockbox to any other better hardware
12:26:02qwmi'm going out for a run soon, and my stopwatch broke last time
12:26:11Unhelpfuland, given the nature of the patch, it's very likely something i *didn't* change that's the problem.
12:26:33 Join PaulJam [0] (
12:26:44qwmso i think i'll use my ipod, with rockbox... my build is a year old. has any significant changes happened to the stopwatch application? is it reliable for a 1h run?
12:27:10bluebrotherI don't think there were changes to stopwatch at all (from a functional side of view)
12:27:19 Quit igwinnimma (Read error: 110 (Connection timed out))
12:27:30bluebrotherat least no changes I can remember. Except plugin API adjustments of course.
12:27:33qwmwhat about the interface? it's not very pretty. hehe.
12:28:16bluebrotherfeel free to improve it ;-)
12:28:26PaulJamwhat makes you think it could be unreliable?
12:28:50*linuxstb has never used it, but if it hasn't been touched for years, then it's probably reliable...
12:28:52qwmPaulJam: i don't know. having my stopwatch break during one of my best runs in a while has made me.. paranoid.
12:28:55bluebrotherits ugliness? ;-)
12:28:58casainhobluebrother: do you know Olimex? had you bought any board from them? what opinion do you have?
12:29:24qwmwhat's the point of running if you don't get to see a nice time
12:29:48kadobanstopwatch doesn't look bad to me...not sure what i'd even do to make it prettier
12:30:08bluebrothercasainho: I don't know them directly. I just remember this guy I was studying with having repeatedly issues with an AVR JTAG bought at Olimex, while I had the luxury of an Atmel one that was working flawlessly as long as I used it.
12:30:19bluebrotherplus, I think their cases look crappy.
12:30:59bluebrothermaybe add the statusbar, a title line and possibly a menu (and option to save the results to a file)?
12:31:31casainhobluebrother: okok :-)
12:31:50casainhobluebrother: what do you mean by "their cases"?
12:32:02bluebrotherbut that's just my personal opinion −− can't really comment on their quality as I never "really" used their hardware.
12:32:23bluebrotherthe cases e.g. the AVR JTAG comes with. Looks kinda like homebrew.
12:33:23qwmoh well... thanks for your guys' insight.
12:33:28bluebrotherOTOH the AVR dragon I use has no case at all :P
12:33:28qwmwish me good luck. ;)
12:33:35bluebrotherqwm: good luck!
12:33:39qwmthanks. :)
12:33:42casainhobluebrother: ah, okok :-) −− well, again, they have cheap hardware, and again, is important for our project. I am using their ARM-USB-JTAG debugger and is good - at least is valuable making the debug :-)
12:34:20bluebrotherhaving a JTAG is always good −− can save hours and hours of work.
12:34:28casainhobluebrother: I also use some Atmel tools at work... and I hate the fact that there is no Open hardware debugger :-(
12:34:54casainhobluebrother: I am prefering now ARM...
12:35:57casainhoto my Open hardware projects I think I will stick with ARM and forget AVR −− I remember that at beginning of this project I wanted to use AVR32 :-)
12:39:10casainhobluebrother: ah, okok, but I am using Tiny AVR wich don't have JTAG, just DebugWire...
12:39:22*bluebrother wants the Makefile recreate itself when configure changed
12:39:31casainhobluebrother: and looks there is no Open hardware debugger for DebugWire
12:39:39Zagorbluebrother: uh, it does
12:39:42 Nick Bensawesome is now known as Bensawsome (n=Bensawso@unaffiliated/bensawsome)
12:39:56bluebrotherZagor: really? It doesn't re-run configure here.
12:40:32bluebrotheror is this tree that outdated? Hmm.
12:40:32Zagorbluebrother: oh, you mean like that. yeah that could be done. I don't much see the point though.
12:41:09bluebrotherwell, I ran into the issue that my other box has a somewhat outdated source tree, but I use it for building h100 every now and then.
12:41:26bluebrothermanually re-running configure is a bit tendious. qmake does that automatically ;-)
12:41:36bluebrotherso I guess I'm somewhat spoiled
12:42:39bluebrothermaybe I should give it a go trying to scratch up a patch
12:43:08Zagoruh, no it doesn't. you may have added rules for it do so, but that's not what I call "automatically"
12:43:33Zagorbut yeah, I agree it would be a good thing
12:47:26Zagorsince you've done it for qmake already, surely a patch for root.make is a quick job?
12:48:23 Quit japc (Read error: 60 (Operation timed out))
12:48:37bluebrotherqmake does that itself, nothing I had to do anything about
12:49:24bluebrotherbut I bet it isn't that hard anyway. The only thing that might need some more adjustments is configure
12:49:35*bluebrother puts this on todo list
12:49:42Zagorit reruns configure with the proper command-line parameters, automatically? somehow I find that hard to believe.
12:50:24bluebrotherqmake creates a Makefile that depends on the pro file, and if that changed it re-runs qmake with the same arguments that have been used during the first run. Quite neat.
12:50:57 Join Minthe [0] (
12:51:04bluebrothermake does support recreating its main Makefile. Read that in the documentation a bit ago
12:51:12Zagorso basically you have entered the command-line parameters into a file?
12:51:14Zagorof course it does
12:51:22Zagorwe already to that today
12:51:58***Saving seen data "./dancer.seen"
12:52:23Zagorbut that is done manually? (specifying the parameters)
12:54:01bluebrotherwith qmake? Only if you want to specify options (like -config silent)
12:54:25ZagorI mean the parameters to ../tools/configure
12:55:54bluebrotherno, I still use the menu. But IMO the resulting options should get saved
12:56:55bluebrotherwhere are additional choices stored? I'm currently looking at the Makefile of a h120 build with rtc mod but can't find any pointer to the rtc enabled
12:57:13*bluebrother spots autoconf.h :)
12:58:10Zagorconfigure should create a CONFIGURE_OPTIONS variable in Makefile that contains the right options to recreate the same file
12:58:32bluebrotheryep, that's what I thought too
13:02:58 Quit Minthe ("Leaving...")
13:05:05 Quit linuxstb (Read error: 113 (No route to host))
13:06:46B4gderand a make target that checks if the makefile was made with an older configure, and then re-run configure with that set of options...
13:07:17ZagorMakfile: ../tools/configure
13:07:27Zagor ../tools/configure $(CONFIGURE_OPTIONS)
13:07:28bluebrotherMakefile: $(TOOLSDIR)/configure
13:07:40 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:07:52B4gdertrue, that should work
13:08:33bluebrotherthe only thing that needs improvements is configure −− afaics it doesn't handle the (a)dvanced entries by command line yet
13:08:51B4gderwasn't there a patch doing that?
13:09:28bluebrotherit does handle command line arguments for the other selections like model and memory which was a patch IIRC.
13:09:49bluebrotherhaven't seen any that also takes the advanced menu into consideration. Will check later.
13:09:55B4gderah, right. It doesn't cover it all
13:14:13 Part B4gder
13:23:09 Join kugel [0] (n=chatzill@unaffiliated/kugel)
13:28:21 Join AndyI [0] (i=AndyI@
13:29:26casainhoI havde my LCD drivers ready. I would like to wrtie a patern on LCD... how can I write on lcd_framebuffer? - I am getting this error: error: ‘lcd_framebuffer’ undeclared (first use in this function)
13:32:59mcuelenaerecasainho: did you do #include "lcd.h" ?
13:33:20casainhoyes, I did
13:35:29mcuelenaerecasainho: can you pastebin a 'svn diff'?
13:36:49casainhoI will :-)
13:37:00 Join kachna|lappy [0] (
13:39:25 Nick fxb is now known as fxb__ (
13:41:55 Quit AndyIL (Read error: 110 (Connection timed out))
13:45:46 Quit linuxstb (Read error: 110 (Connection timed out))
13:48:02casainhomcuelenaere: here is the patch file:
13:49:14mcuelenaerecasainho: why didn't you defined HAVE_LCD_BITMAP?
13:49:39casainhoah... okok :-)
13:50:06mcuelenaereand you haven't defined LCD_PIXELFORMAT either
13:50:15casainhohow can we use 0b00000001 on gcc?
13:50:59 Part Anges
13:50:59mcuelenaereI don't think it's possible to do that, use 1 or 0x1 instead
13:51:44casainhookok :-)
13:53:35kugelbluebrother, Bagder: no it doesnt. At least the initial patch didn't do it
13:53:37casainhomcuelenaere: thank you
13:53:49mcuelenaerecasainho: did it fix the error?
13:54:03casainhoyes, it did :-)
13:59:00kugelUnhelpful: I thought there's a real name policy for commit messages
14:00:08 Join aurix_lexico [0] (
14:03:24 Join massiveH [0] (
14:03:25 Quit massiveH (Client Quit)
14:04:32 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
14:05:57 Quit CaptainKewl (Read error: 104 (Connection reset by peer))
14:06:07mcuelenaerecasainho: why do have several '../firmware/' in that patch?
14:06:18 Join CaptainKewl [0] (
14:06:41casainho../firmware ? to include files?
14:06:54mcuelenaere(and you shouldn't include your eclipse settings)
14:07:15mcuelenaereyes, for example in firmware/SOURCES: ../firmware/target/arm/at91sam/rockboxplayerlittle/timer-rockboxplayerlittle.c
14:07:31 Join flydutch [0] (
14:08:12PaulJamhmm, is it intended, that the controls in the quickscreen are inversed for numeric lists like for example brightness (UP decreases the value and DOWN increases it)?
14:08:38casainhomcuelenaere: I don't know, but I think I must include on that way, no?
14:08:58mcuelenaereif all others can do it without those relative path, why can't you?
14:09:05casainhomcuelenaere: and the Eclsipe settings are for others developers... I hope it help
14:09:29mcuelenaereok, but could you not 'svn add' them then?
14:09:31casainhomcuelenaere: eheh, so, I will start removing that paths :-)
14:10:09casainhomaybe when I build the patch for you, but that pacth was mainly for our groups of developers...
14:10:26casainhoI just used the same patch...
14:10:43casainhookok, next time i will not add that files
14:11:10mcuelenaereah ok
14:13:32 Quit kachna|lappy (Read error: 113 (No route to host))
14:18:02 Join LambdaCalculus37 [0] (i=44a04303@rockbox/staff/LambdaCalculus37)
14:19:31mcuelenaerecan a Rockbox LCD driver yield()/sleep() (in lcd_update()) or is there a possibility that it's called within interrupt context?
14:23:38 Join tvelocity [0] (
14:28:43 Join Schmogel [0] (
14:31:18casainhomcuelenaere: I think it can called within interrupt context. For example, my lcd_update call SPI drivers which can use interrupts.
14:31:25J-23casainho: do you have any hardware for RockboxPlayer?
14:32:24mcuelenaerecasainho: I meant whether lcd_update() gets called from within interrupt context
14:33:12kugelmcuelenaere: I don't think it does
14:33:25mcuelenaereok, so it's safe to yield() in it?
14:33:45kugelwhy would you yield?
14:33:54mcuelenaerebecause of a DMA transfer
14:33:59mcuelenaerefrom the CPU to the LCD controller
14:34:11mcuelenaereand it isn't actually sleep()'ing but, wakeup_wait()
14:35:19kugelmcuelenaere: I don't quite understand. You want to yield in the function that invokes the transfer?
14:35:59casainhoJ-23: yes, I have :-) −− but please don't laugh :-) :-)
14:36:27J-23looks interesting ;p
14:36:43mcuelenaerekugel: yes, the function initiates the DMA transfer and as this is asynchronous to the CPU, it can yield() and do other tasks
14:36:49 Join mib_528f61m6 [0] (i=cf6be8f5@gateway/web/ajax/
14:37:10casainhoJ-23: as you can see, I just have the LCD for now, I am working on the drivers for it...
14:37:31kugelmcuelenaere: well, I don't see a reason why you couldn't yield
14:37:32J-23isn't it small?
14:38:00casainhoJ-23: oh yeah, sery small, very portable ;-)
14:38:03mcuelenaerekugel: if you're in an interrupt, there could be some trouble
14:38:22J-23casainho: and very hard to watch movies/browse photos on it ;p
14:38:50casainhoJ-23: ah, you mean, the LCD? yes, is the cheapest one and most availably LCD... it's used in almost every Open Hardware projects that uses a LCD
14:39:04kugelmcuelenaere: yes I know, but lcd_update isn't called in ISRs
14:39:16casainhoJ-23: you know, the objective is to make a audio player and recorder, no multimedia
14:39:42kugelmcuelenaere: look at lcd-Xbit.c, there are just some wrappers for lcd_update_*, e.g. the viewport functions
14:39:43casainhoJ-23: BUT, this is a kind of a temporary prototype, it's for being cheap, not as the final
14:39:53kugelthat's what the app code calls
14:40:04kugelnothing with IRQ as far is I can see
14:40:09mcuelenaereok, that's what I wanted to know
14:40:41J-23hm, is there any full list of hardware used to build it?
14:42:57casainhoJ-23: yes, more or less, see here:
14:43:16amiconncasainho: If your LCD only does RGB444, you should use that pixel format in rockbox as well for performance
14:43:46amiconn(well, actually XRGB4444 with X being unused bits)
14:43:49casainhoamiconn: sorry, don't really know that do you want to say...
14:45:45casainhocan someone please help me on this error:
14:46:06*amiconn once considered going RGB444 for the c200, back when it was unclear whether the transfer can be sped up enough to be usable for video etc
14:47:20mcuelenaerecasainho: you seem to have SYSFONT_WIDTH & SYSFONT_HEIGHT defined in your config-..h ?
14:47:59 Quit HellDragon (Remote closed the connection)
14:48:23 Join HellDragon [0] (
14:48:42 Quit Llorean ("Leaving.")
14:48:43casainhomcuelenaere: yes: #define SYSFONT_WIDTH 10
14:48:45casainho#define SYSFONT_HEIGHT 10
14:48:54mcuelenaereyou shouldn't do that
14:49:11 Join Llorean [0] (
14:49:18mcuelenaerethe build system/font.h takes care of that
14:51:11 Join itcheg [0] (i=41d59de2@gateway/web/ajax/
14:51:24casainhomcuelenaere: okok. that warnings don't appear any more
14:52:02***Saving seen data "./dancer.seen"
14:53:19casainhomcuelenaere: did you saw the picture of my hardware?
14:54:27mcuelenaeregevaerts: sending 13bytes of data through EP1 works, sending 36bytes doesn't. any tips? :)
14:55:09mib_528f61m6If the missing .dll problem is fixed in 1.0.9 I think the warning should be removed or at least modified on the frontpage.
14:55:21casainhomcuelenaere: :-)
14:55:33Zagoris it fixed?
14:56:07 Nick mib_528f61m6 is now known as MarcGuay (i=cf6be8f5@gateway/web/ajax/
14:56:32gevaertsmcuelenaere: yes. Use a hardware tracer. That way you can see what actually happens.
14:57:04gevaertsI should be able to get it to the post office tomorrow
14:58:09MarcGuayZagor: Looks like it.
14:58:34mcuelenaeregevaerts: how do you connect it to a pc? also through USB?
14:58:55Zagorfront page updated
15:02:04casainhowhy I am getting this error? : /home/cas/Documentos/rockbox_player/rockbox/firmware/panic.c:88: undefined reference to `system_exception_wait'
15:02:30mcuelenaerecasainho: that's due to a recent commit, you'll need to implement system_exception_wait()
15:02:35 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
15:02:46mcuelenaerecasainho: see;revision=19716
15:04:00casainhomcuelenaere: I didn't understand in where should I implement it....
15:05:09mcuelenaerecasainho: then just use for now
15:06:25mcuelenaerewhen you know what GPIO channel is mapped to what button, you can replace it with a while() loop that exits() upon pressing a button
15:06:41 Join Anges [0] (
15:06:46gevaertsmcuelenaere: yes. It works best with a second PC, but you can use the same PC as well
15:07:14mcuelenaeregevaerts: and a second pc on the same pc aka a VM, would that work too?
15:07:52casainhomcuelenaere: thanks. I putted it on system-*.c and now build without any error :-)
15:08:04gevaertsmcuelenaere: that should work, but the reason a second PC works best is that you have separate USB busses, so it doesn't sniff its own traffic.
15:08:39gevaertsmcuelenaere: the software has a setting to tell it to ignore its own traffic though. It's just a bit less convenient
15:08:50mcuelenaerehmm but the device only has 2 USB ports, doesn't it? so I presume it has an internal buffer which stores the USB traffic?
15:09:06mcuelenaereand then afterwards pushes this to the PC?
15:09:08mcuelenaereor am I wrong?
15:09:56gevaertsThat's what it does, yes. It has a set off pass-through ports (one A and one B) through wich you connect the device to test, and one port to connect it to the controlling PC
15:10:44mcuelenaereso why should it ignore its own traffic?
15:11:04mcuelenaerebecause it doesn't generate any while the to-be-tested device is busy doing it's own
15:11:17gevaertsThere is communication while capturing as well, mostly status info I guess
15:11:22mcuelenaereah ok
15:11:40mcuelenaereand that doesn't mess up the original USB transfers?
15:11:49mcuelenaereor does it behave like a usb hub then?
15:12:10gevaertsNo and no :)
15:12:20mcuelenaereok :)
15:12:43gevaertsUSB outgoing traffic is broadcast to all ports, and devices ignore anything that's not for them
15:13:32mcuelenaerethat sounds a bit risky to me, as these devices are supposedly to be tested so they could not ignore that data (but that doesn't apply in our case)
15:14:30gevaertsThis complexity is why using separate PCs (or USB controllers) is recommended :)
15:14:59 Join evilnick [0] (i=0c140464@gateway/web/ajax/
15:15:57gevaertsAnyway once you start using it things will become clearer
15:16:31*mcuelenaere wonders how gevaerts uses it, as the software seems to be Windows-only
15:16:54gevaertsusing Windows? :)
15:18:51 Join japc [0] (n=japc@
15:25:22 Join tyfoo [0] (
15:25:58 Quit mcuelenaere ("gtg")
15:33:34 Part LinusN
15:39:42 Quit petur ("drink time :)")
15:40:08 Join Jaykay [0] (
15:41:16 Quit MarcGuay (" ajax IRC Client")
15:44:16 Quit linuxstb (Read error: 60 (Operation timed out))
15:57:37 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
16:08:03 Join Asherael [0] (
16:08:20Asheraelhow do I determine if I have 3.0 or 3.1 installed?
16:09:13LloreanCheck the version number, I believe under the "System" option?
16:09:41 Quit Jaykay (Read error: 110 (Connection timed out))
16:10:29AsheraelI have the 11 digit version number
16:11:28LloreanIt doesn't have a -3.0- or a -3.1- in it?
16:12:12 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
16:13:15LloreanThen you have neither 3.0 or 3.1
16:13:21LloreanYou have one of the many development builds
16:13:52Asheraeloh, thanks
16:15:03 Join TheSphinX^ [0] (
16:20:14 Join Jaykay [0] (
16:24:20 Join {phoenix} [0] (
16:25:05 Quit linuxstb (Read error: 110 (Connection timed out))
16:25:06 Join jgarvey [0] (
16:25:09 Join MethoS- [0] (
16:27:45 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.5/2008121622]")
16:30:26 Join linuxstb [0] (
16:41:26 Join toffe82 [0] (
16:46:47 Join kachna [0] (
16:52:05***Saving seen data "./dancer.seen"
17:04:25 Join likemindead [0] (
17:05:45 Part likemindead ("Making like a tree...")
17:05:59 Quit Jaykay (Read error: 110 (Connection timed out))
17:08:11 Quit Zagor ("Client exiting")
17:08:29 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
17:08:46 Nick fxb__ is now known as fxb (
17:22:52 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
17:23:28 Quit matsl (Read error: 110 (Connection timed out))
17:25:55 Quit blahrus ("Ex-Chat")
17:31:48 Quit PaulJam (Read error: 113 (No route to host))
17:33:05 Quit MethoS- (Read error: 131 (Connection reset by peer))
17:33:18 Join miepchen^schlaf [0] (
17:33:58 Join MethoS [0] (
17:35:37 Quit miepchen^schlaf (Client Quit)
17:37:23 Join miepchen^schlaf [0] (
17:39:26 Nick krazykit` is now known as krazykit (
17:45:51 Join nibbler [0] (
17:46:16 Quit troy__ (Read error: 104 (Connection reset by peer))
17:50:01 Join kugel [0] (
17:54:30 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
17:59:36 Quit itcheg (" ajax IRC Client")
18:00:43 Quit Xerion (" ")
18:08:43 Nick fxb is now known as fxb__ (
18:14:36 Quit timc`` (Read error: 104 (Connection reset by peer))
18:14:52 Join timc`` [0] (n=aoeu@
18:18:48 Join karashata [0] (n=karashat@
18:21:00 Join bertrik [0] (
18:22:58 Join sarixe [0] (
18:50:55 Quit Darksair ("Do you hear that? This is the sound of inevitability. This is the sound of your death, Mr. Anderson.")
18:50:58 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
18:52:06***Saving seen data "./dancer.seen"
19:00:02 Join gregorovius [0] (
19:01:07 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
19:01:14 Quit MethoS (Remote closed the connection)
19:07:40 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
19:10:42 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
19:13:07 Join Thundercloud [0] (
19:13:27 Join gregzx [0] (
19:17:30 Quit TheSphinX^ ("XChat@Linux")
19:19:06 Join Xerion [0] (
19:23:15 Join fml [0] (n=4fd3c146@gateway/web/cgi-irc/
19:24:42 Join z35 [0] (
19:25:32fmlMarcGuay: Hello. With r19734 you did what I have not intended in FS #9769 and did not what I have intended. So why the credit?
19:27:18 Quit tyfoo ("Carpe diem")
19:29:04 Join BigBambi [0] (n=alex@
19:30:53 Join tyfoo [0] (
19:34:15 Quit fml ("CGI:IRC (EOF)")
19:35:54 Join Jaykay [0] (
19:38:19 Quit BigBambi (Read error: 113 (No route to host))
19:42:42 Join BigBambi [0] (n=alex@
19:45:23 Join tessarakt [0] (
19:45:56 Quit BigBambi (Remote closed the connection)
19:48:02 Part toffe82
19:49:15 Join BigBambi [0] (
19:55:51saratogasomething looks screwy to me in the mp4 seeking code
19:56:15saratogawe allocate more seek points then we have frames in the file
19:58:34 Join Zambezi [0] (i=stolgfor@
19:58:57 Quit japc (Read error: 145 (Connection timed out))
19:59:04 Part Asherael
19:59:05kugelsaratoga: hey, I have a crappy sync of your rockbox-as-an-app patch (so it at least builds)
19:59:47saratogakugel: oh wow
20:00:00kugelI find some parts of it very weird. It looks like you could decide between adding a new target, or extending the sim
20:00:32saratogakugel: the idea was to add another target, but unfortunately the target i wanted to add ended up not being an option, so it mostly just duplicates sim code into the target tree
20:00:45saratogai was never very happy with it actually
20:00:59saratogawhat are you working on?
20:01:25kugelnothing :) just sync'd it to see how it is
20:02:41saratogai'm trying to get the aac memory usage under control so it can be used on low mem targets
20:02:57 Quit jhulst (Read error: 145 (Connection timed out))
20:03:30kugelI think it could/should be done like this: rockbox app is the default sim. target-sims just pass their configs, which will modify rockbox app so that it matches the target a bit more. This will probably get the sims a bit farer away from the targets, but it will force the rockbox app to kept working
20:03:57BagderI disagree with that approach
20:04:30gevaertsThe first purpose of the sim is to be close to the actual target IMHO
20:05:15saratogathe app idea is weird because it starts out as just the sim, but if you ever did an actual port you'd quikcly need to add target specific items too
20:05:19kugelwell, I got the impression that this isn't enforced anymore recently, and the sims are practically already very far away
20:05:30gevaertsAnd the purpose of the app is to be used. Once it's really used, it will keep working
20:05:43Bagderand added to the build table etc
20:05:52kugelplus, the target sims pass their config-<target>.h, so I don't really expect so much changes
20:06:20Bagderkugel: in what way is that not "enforced" anymore?
20:06:36saratogarealistically a mature rockbox as an app port would probably not be much like a sim build, since it'd still have most or all of the drivers a firmware target has
20:06:51saratogajust they would wrap host os functions
20:08:48kugelBagder: what would actually change with that approach? the sim mostly doesn't use the target-specific firmware/ code anyway, and the apps/ code will be the same for rockbox app
20:09:07saratogathe mp4 parser has me really confused
20:09:53Bagderkugel: it was you who said they would be further away and I objected
20:10:18saratogait seems to allocate ~5KB a minute worth of RAM just for a table that contains sizes needed for each frame when it calls ci->request_buffer
20:10:24BagderI want the "app" version to be as an actual port
20:10:28saratogawhy not just always request the maximum needed size
20:10:44LloreanThe "app" version should actually be several ports.
20:11:00BagderLlorean: indeed, I mean an app for a specific OS/purpose
20:11:02kugelI said probably, so it wasn't definite. but the more I think about it, the less probable it is
20:12:40 Quit Jaykay (Read error: 110 (Connection timed out))
20:12:56kugelrockbox-as-an-app was a gsoc project, right?
20:13:30kugelBagder: that's what I would want too
20:13:34 Quit Horscht (Read error: 54 (Connection reset by peer))
20:13:47*gevaerts thinks that RaaA will probably be used a lot on systems with touchscreens
20:15:33 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
20:17:53 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
20:21:46 Quit ze (Read error: 113 (No route to host))
20:29:58 Nick Bensawsome is now known as DOMO (n=Bensawso@unaffiliated/bensawsome)
20:31:25 Join idshark [0] (
20:31:32 Join Jaykay [0] (
20:31:45 Nick DOMO is now known as Bensawsome (n=Bensawso@unaffiliated/bensawsome)
20:52:10***Saving seen data "./dancer.seen"
20:53:20 Join Zagor [242] (
20:56:14 Quit saratoga ("CGI:IRC (Ping timeout)")
20:57:06 Quit Xerion (" ")
20:57:36 Join ze [0] (i=ze@
20:59:51 Join m0f0x [0] (n=m0f0x@
21:09:54 Quit BigBambi (Read error: 113 (No route to host))
21:10:13 Join pyro_maniac [0] (n=jens@
21:12:30 Quit Thundercloud (Remote closed the connection)
21:14:24 Join itcheg [0] (i=62db4767@gateway/web/ajax/
21:14:54 Join mib_zixhxrsj [0] (i=cf6be8f5@gateway/web/ajax/
21:14:57 Join blahrus [0] (n=blahrus@
21:15:20 Nick mib_zixhxrsj is now known as MarcGuay (i=cf6be8f5@gateway/web/ajax/
21:16:47MarcGuayfml: Hi. I believe the changes I made did what your patch did and more. However, the spacing changes don't seem to show up in the diffs presented by the website.
21:17:23 Quit tessarakt (Read error: 110 (Connection timed out))
21:17:50 Join tessarakt [0] (
21:18:06 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:20:31ZagorMarcGuay: correct, the diffs don't show space changes
21:21:14Bagdernot the colored ones, you can select a unidiff to see them
21:21:32Zagorright, there are several options
21:24:02MarcGuayfml: See here for proof: Thanks for the tips Zagor & Bagder.
21:25:50*MarcGuay notes a case where proper punctuation deters effective communication (link breaks because of the period)
21:26:33Zagornot in xchat, interestingly
21:26:45pixelmadepends on the cleverness of the IRC client
21:27:18 Join Seed [0] (
21:27:19ZagorI guess periods at the very end of urls are pretty uncommon :)
21:28:59 Join petur [50] (n=petur@rockbox/developer/petur)
21:29:34*Zagor wonders where the last 38KB of pcmbuffer disappears
21:30:21 Join Aurix_Lexico1 [0] (
21:31:13Zagorbuffer_size:176400 unplayed_bytes:103424 free:4608
21:31:29 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
21:32:21Zagorand that doesn't add up to those 38KB even :)
21:33:06 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
21:35:05kugelMarcGuay: BTW: I thought there's a real name policy for commit messages
21:35:21bluebrothersince when?
21:35:45ZagorI'd say it's more of a tradition than a policy.
21:35:51 Quit tessarakt (Connection timed out)
21:36:04kugelI don't know how official it is, but I noticed several times that it was "complained" about using the nick name
21:36:20 Join tessarakt [0] (
21:36:33ZagorI personally think real names are better in the commit logs.
21:36:38kugele.g. back when funman committed a patch of mine, and wrote kugel in the commit message
21:37:46robin0800real names are needed to edit the wiki
21:37:50bluebrotherit's rather easy to match the nickname -> real name on long(er) term contributors. I agree that it would be nice to use the real name to cut the need of resolving that later :)
21:37:56 Quit itcheg (" ajax IRC Client")
21:40:35 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
21:41:44*kugel can't find it in the logs, but I am sure he was asked to use real names
21:43:58 Join itcheg [0] (i=62db4767@gateway/web/ajax/
21:44:49 Quit karashata ("G'bye everyone!")
21:45:14 Join nplus [0] (n=nplus@141.25.Globcom.Net)
21:45:39Zagorkugel: please do, even if don't have a written policy about it
21:46:57kugelI should use real names in commit messages? :)
21:47:04 Join akur [0] (
21:47:09 Quit jhulst (Connection timed out)
21:47:10MarcGuaykugel: Clicking on the FS link will show you his real name quite quickly... And the commit itself is under my name, which is real.
21:48:26 Part akur
21:49:21*bluebrother is confused about who was asked to use real names. Kugel?
21:49:58kugelI referred to funman, not sure who Zagor referred to
21:50:32 Quit jhulst_ (Read error: 60 (Operation timed out))
21:50:37Zagorkugel: :)
21:51:08bluebrother* kugel can't find it in the logs, but I am sure he was asked to use real names <- implies that kugel was asked ...
21:51:46kugel"but I am sure he was asked" he != I
21:52:09kugelthe second part should've not been /me'd I suppose ;)
21:52:31 Quit robin0800 (Read error: 54 (Connection reset by peer))
21:52:32MarcGuayThe third person is a tricky thing.
21:52:32MarcGuayA sign of severe mental problems, apparently. :)
21:52:50 Join robin0800 [0] (
21:53:35*bluebrother wonders why the second part shouldn't be 3rd person, especially as it even isn't a separated sentence
21:54:10kugelbecause there was an "I" in there
21:54:20kugelthis tends off-topic..
21:54:31bluebrotherindeed, and I still disagree :P
21:55:03 Quit Nico_P (Remote closed the connection)
21:56:03 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
21:58:39 Join Thundercloud [0] (
21:59:38 Quit LambdaCalculus37 (" ajax IRC Client")
22:02:40MarcGuayCan someone confirm that the remaining "Settings which don't work as expected" on the QuickScreen wiki page has been fixed? (Status Bar - the quickscreen and menu doesn't get immediately redrawn)
22:03:11MarcGuayI've been using this regularly and haven't had any problems.
22:05:12 Quit tvelocity (Remote closed the connection)
22:07:20Jaykayi suggest committing of and to reduce those really old patches. gevaerts said already he'd commit them but didnt do it because of the 3.1 freeze...
22:07:52Jaykayif you dont want to add me to the credits for this snake-patch, you dont need to :)
22:11:05Jaykaysilence silence silence...
22:15:43unstableJaykay: What device do you own?
22:16:26MarcGuayEven committing simple patches requires time and effort.
22:18:57Jaykayi could help you with it. ha.
22:19:05Jaykayunstable: is it important?
22:20:36unstablecurious what your patches are for
22:20:59unstableI have the e280 also, but I have the v2. You have v1?
22:22:21Jaykayi thought it was obvious what the patches do
22:22:43MarcGuaySome sweet and rare appreciation for the manual:
22:27:52 Join BigBambi [0] (
22:31:50 Quit flydutch ("/* empty */")
22:32:39 Quit ze (Read error: 113 (No route to host))
22:34:16 Join jkeller [0] (n=1815ce70@gateway/web/cgi-irc/
22:34:44evilnickhi there
22:34:47jkellerI've been trying to install Rockbox on my iPod
22:34:57jkellerbut I put the .rockbox directory on the ipod volume, right?
22:35:05jkellerwhen I boot, the firmware says it can't find "rockbox.ipod"
22:35:15evilnickWhat OS are you on?
22:35:19jkellerMac OSX
22:35:29jkellerI used the mbr conversion to convert to Fat32
22:35:41evilnickAre you putting the ".rockbox" directory on the root of the ipod?
22:35:48 Join ze [0] (i=ze@
22:35:49jkellerat least I think I am.
22:35:55jkellerat least I think I am.
22:36:02jkellerI open the ipod in disk mode
22:36:11jkellerand put .rockbox in the /Volumes/IPOD directory
22:36:15jkelleronce it is mounted
22:36:27 Quit mcuelenaere (Read error: 113 (No route to host))
22:36:33jkellerit will still boot to the apple firmware if I turn the hold switch the other way
22:36:58jkellerthe firmware says partition 1
22:37:02jkelleron the black screen
22:37:13evilnickHmmm, I'm not (at all) familiar with OSX - can you dual boot into Windows? Or perhaps wait until someone more knowledgeable wakes up
22:37:18jkellerbut it's on 20mb, so do I need to put rockbox on that partition, rather htan the 80gb
22:37:24jkellerI can wait
22:37:27jkellercause I don't have windows.
22:37:29 Join DrMoos [0] (
22:37:35 Quit moos (Read error: 104 (Connection reset by peer))
22:37:43 Nick DrMoos is now known as moos (
22:38:15domonokyjkeller: .rockbox has to go to the second partition..
22:38:47jkellerI put it there
22:38:50domonokyyou normally also can not write to the first partition on ipods (no filesystem)
22:38:52 Join japc [0] (
22:38:55jkellerit's not being recognized
22:39:00jkellerI didn't think I could.
22:39:06evilnickjkeller: Or you could try the rockbox utility:
22:39:25jkellerI tried that, and even though ipodpatcher recognizes this as a winpod, the utility does not.
22:39:54jkellerI used the exact instructions for patching the mbr from the wiki
22:40:02jkellerand iTunes sees it as a winpod
22:40:04 Quit Zagor ("Clint excited")
22:40:40domonokyplease make sure you really have .rockbox/rockbox.ipod in your root dir..
22:41:35jkellerI do.
22:41:44jkellerthe iPod is mounted in diskmode as /Volumes/IPOD
22:41:50jkellerand it has a .rockbox folder
22:41:55 Quit bmbl ("Woah!")
22:41:55jkellerand inside it has a 604k file called rockbox.ipod
22:42:21domonokysounds right.
22:42:51jkellerI think the problem is that I didn't use Windows to make this a winpod
22:43:02jkellercause I downloaded the MBR file.
22:43:18jkellerto change it, since I don't own a windows machine
22:43:19domonokyno, the manual conversion should also work.
22:43:45jkellerI do know the rockbox utility doesn't recognize it as a winpod
22:43:55jkellerbut I reformatted and tried a second time just to make sure
22:44:13domonokywhich is strange, it uses the same code as ipodpatcher for the bootloader install.
22:44:14evilnickjkeller: Did you reformat using iTunes?
22:44:18jkelleris there a log file or something it keeps?
22:44:24jkellerand then changed it to a winpod again
22:44:29jkellerusing the steps in the guide
22:44:43jkellerIt's a 5.5th gen iPo
22:44:50domonokywhat does autodetection in rbutil say ?
22:44:52jkellerand I used disk utility to format the partition
22:45:01jkellera 30Gb 5th gen
22:45:14jkellerwhich the pdf manual said it can't detect if it's an 80gb yet
22:45:15domonokyare you sure the partition is now formated fat32 ?
22:45:29jkelleraccording to diskutil it is
22:45:57jkellerit says it's an MS-DOS (FAT32)
22:46:39 Quit XavierGr (Nick collision from services.)
22:46:44jkellerhang on
22:46:49jkellerI'm going to get the full error message
22:46:50 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
22:50:03jkellerit lists Partition 1
22:50:09jkellerbut it doesn't appear to list partition 2
22:50:15jkellerwhich is where the .rockbox stuff is
22:50:20jkellerwould that be a problem?
22:50:24jkellerlike, should it be listing both
22:50:39evilnickjkeller: I would assume so
22:50:43domonokyi dont know what you use to list the partitions... so i dont know :-)
22:50:51jkellerno I mean
22:50:54jkellerthe rockbox error
22:50:59jkellerwhen I try to boot into rockbox
22:51:08jkellerit has a black screen with information
22:51:10jkellerabout the device
22:51:11domonokyplease give the exact error messages.
22:51:16jkellerhang on
22:51:25jkellerRockbox boot loader Version 3.0 IPOD version: 0x000B0011 Partition 1: 0x0B 19045 MB Loading Rockbox... Error! Can't load rockbox.ipod: File not found Hold MENU+SELECT to reboot then SELECT+PLAY for disk mode
22:51:31jkellerthat's it without the blank lines
22:51:38jkeller* I mean line breaks*
22:52:14***Saving seen data "./dancer.seen"
22:53:01 Part Aurix_Lexico1
22:53:02domonokythats ok, he first partition is without filesystem so rockbox ignores it. It correctly finds the 20gig partition, but fails to find the files.
22:53:36domonokyi still suspect that somehow the files are not on the correct place. maybe the moutpoint was not correct (maybe a left-over folder? )
22:54:03jkellerI have an 80gb ipod
22:54:05jkellerand that's a 20 mb partion
22:54:30domonoky19045 MB > 20MB
22:54:38jkelleryou're right xG
22:54:52jkellerbut why would it only find a 20Gb partition?
22:54:56jkellerinstead of an 80gb one.
22:54:57domonokymaybe you used a mbr from a 20gig ipod ?
22:55:03jkellerI'll check again
22:55:09jkelleralso, I didn't use mtools to format
22:55:10 Quit MarcGuay (" ajax IRC Client")
22:55:12jkelleri used the disk util
22:55:15jkellerbut I'll try mtools
22:55:25 Quit jkeller ("CGI:IRC")
22:55:45scorchejkeller: please stop using the enter key as substitute for makes it harder to read and a bit spammy where it does not need to be..
22:57:32 Quit petur ("zzzzz")
22:59:16Jaykayunstable: whats about the two patches?
23:03:02 Quit m0f0x (Connection timed out)
23:13:08 Quit japc (Read error: 110 (Connection timed out))
23:16:04 Quit itcheg (" ajax IRC Client")
23:18:04 Quit gregorovius ()
23:18:33 Quit evilnick (" ajax IRC Client")
23:21:56 Join Jaykay_ [0] (
23:23:07 Join DrMoos [0] (
23:23:24 Quit moos (Read error: 104 (Connection reset by peer))
23:23:41 Nick DrMoos is now known as moos (
23:31:24 Quit Jaykay (Read error: 110 (Connection timed out))
23:32:53 Join Xerion [0] (
23:35:45 Quit bertrik ("Leaving")
23:36:18 Quit linuxstb (Read error: 113 (No route to host))
23:37:46*Bagder wants a commit!
23:39:19Bagdernew magic scripts...
23:40:10kugeltell me more! what are they supposed to do?
23:40:15 Join akur [0] (
23:40:33Bagderdetect red builds
23:41:01kugelyou need to wait for JdGordon though to make sure it works :P
23:42:36 Join fml [0] (n=4fd3c146@gateway/web/cgi-irc/
23:43:37 Quit Jaykay_ (Read error: 110 (Connection timed out))
23:44:49 Part akur
23:46:08fmlErr.. MarcGuay is gone... Anyway: I asked because I couldn't recognize my changes in the diffs displayed in the "normal mode." And (I didn't write it but that was the start of the patch) the main "feature" of the patch was the correct spacing after the macros for the key to press. But now everything is even better than I could dream of! :-)
23:46:39fmlAnd now I'll join MarcGuay, i.e. leave the channel :-) Bye
23:46:42 Quit fml (Client Quit)
23:47:37 Join Beta2K_ [0] (
23:53:13 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
23:56:18 Quit ender` (" I think I remember an episode of MacGyver where he overthrew a violent dictator with a rubber band, 2 bottle caps, and some")
23:57:14 Quit nplus (Remote closed the connection)
23:57:17 Join japc [0] (

Previous day | Next day