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

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

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

#rockbox log for 2009-04-11

00:00:17HillshumI'll do it once I have your name
00:00:41arez-li cannot make a (at) on this vm O.o ... gonna google it and copy/paste xD
00:02:36arez-ldone installing the compiler
00:03:03 Quit {phoenix} (Remote closed the connection)
00:04:42arez-lthe wiki refers to the stable rockbox ,what files should i get for testing?
00:05:05Hillshumwhat patches?
00:05:40 Quit merbanan (Read error: 110 (Connection timed out))
00:05:45arez-lwhat do you mean?
00:06:06 Join DrMoos [0] (
00:06:11Hillshumwhere does it refer to stable rockbox?
00:06:27 Quit moos (Read error: 104 (Connection reset by peer))
00:06:30arez-li think so at least
00:06:34 Quit DrMoos (Client Quit)
00:06:48 Join moos [0] (i=mustapha@rockbox/staff/moos)
00:06:48arez-lwhat files do i need if i want to get your latest work for testing ?
00:07:18Hillshumfirst try compiling an unchanged build to make sure it all works
00:07:43Hillshumand get shared folders up in Vbox
00:08:40arez-lnow iam doing svn co svn:// rockbox
00:08:54Hillshumdid you do that already?
00:09:28arez-lnot done so far
00:10:03arez-lsorry for asking so much but i never compiled before and iam not quite sure about what the wiki says
00:10:27arez-lnow i got /rockbox in my home dir containing several folders
00:10:59HillshumHave you read ?
00:11:23arez-l:D there is a simple version? thank god!
00:11:32arez-lgonna work trough it
00:14:39HillshumArez-l you can skip to
00:16:26Hillshumalso, before you can get Rockbox to work you'll need the bootloader
00:18:05arez-lsry, didnt see your msg, i followed the guide already up to compiling but i get some errors (
00:19:40BigBambiarez-l: after you ran did you put the cross compilers into your PATH as instructed?
00:20:04Hillshumcopy "export PATH=$PATH:/usr/local/arm-elf/bin" to your $HOME/.bashrc file and relogin
00:20:25arez-li copied it but i didnt relog
00:20:27 Join ultrasonic [0] (n=manohar@
00:21:18arez-lO.O ... its already 0:20 here... i need to get some sleep now
00:21:31arez-ltomottow i will try to get a working bootloader
00:21:36 Quit ultrasonic (SendQ exceeded)
00:21:59arez-lif i use a patched firmware, is it still possible to update the OF of the player?
00:22:07 Join ultrasonic [0] (n=manohar@
00:22:19 Quit ultrasonic (Read error: 104 (Connection reset by peer))
00:22:33 Join ultrasonic [0] (n=manohar@
00:22:40Hillshumyou will need to repatch it
00:22:50 Part ultrasonic ("Leaving")
00:23:44arez-lgood night, and thanks for your help
00:24:21 Quit arez-l ("CGI:IRC")
00:24:29HillshumBut 3.01.16 will be the last so I wouldn't worry
00:26:10 Join goffa [0] (n=goffa@
00:27:59***Saving seen data "./dancer.seen"
00:31:19 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
00:31:34 Quit bluebrother ("leaving")
00:38:39 Quit Hillshum (Read error: 54 (Connection reset by peer))
00:40:20 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
00:40:47 Quit ender` (" Why geeks like computers: look chat date touch grep make unzip strip view finger mount fsck more fsck yes spray umount slee")
00:41:31 Quit goffa_ (Read error: 110 (Connection timed out))
00:44:57 Quit Arez ("CGI:IRC")
00:50:49 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
00:51:45 Join Hillshum [0] (n=chatzill@unaffiliated/hillshum)
00:55:28 Join Horscht86 [0] (
00:56:15 Quit faemir ("Lost terminal")
00:56:36 Quit Horscht86 (Client Quit)
00:56:43 Quit bmbl (Client Quit)
00:56:50 Join faemir [0] (
00:59:21 Quit bertrik ("Leaving")
01:03:53Unhelpfu-amiconn: actually, looking at the idct routines in our decoder, they seem to have been adapted from the IJG ones, or at least many of the macros are the same in their names and implementations. perhaps the best way to go is to adapt our existing decoder to process the file in by rows, and then port more idct routines if we want more scaling options.
01:04:08 Nick Unhelpfu- is now known as Unhelpful (
01:05:32 Quit gregsparc (Read error: 113 (No route to host))
01:08:44 Quit Horscht (Read error: 110 (Connection timed out))
01:11:13 Quit midijunkie (Read error: 104 (Connection reset by peer))
01:12:06 Quit flydutch ("/* empty */")
01:12:48kugelUnhelpful: I have audio drop outs in pictureflow on my fuze
01:13:03kugelis pictureflow yield()ing enough?
01:13:14 Quit Horschti (Read error: 110 (Connection timed out))
01:16:28Unhelpfulper slide rendered... you could try moving the yield() in render_slide up a bit. i'd try first moving it up to the top, before any cases where it returns early - this will make it yield per slide even if none of the slide is actually drawn. that will only matter if you've got the number of slides set high enough that it's trying to draw many slides that are not on screen.
01:17:01Unhelpfulyou could move it inside the column render loop, which will make it yield per column, but i really think that might be excessive.
01:17:39Unhelpfulit could also be that it's the load thread that's not yielding enough, but that yields per slide loaded, and i rather doubt that we really need to yield in between reads while loading a slide.
01:17:41 Quit robin0800 (Read error: 104 (Connection reset by peer))
01:18:33*Llorean adds a "Third Party" section to the forums.
01:19:20 Join timc`` [0] (n=aoeu@
01:19:44BigBambiLooks good
01:20:29kugelUnhelpful: only drop outs while drawing
01:20:40BigBambiI also like the Getting Started change to Starting Development
01:20:45LloreanIf the Third Party one gets particularly busy, I'll add sub-boards like in Manual Installation for categories of tools, like "Video Conversion" etc.
01:20:54Unhelpfulyou mean while scrolling?
01:21:36LloreanBigBambi: Yeah, I decided "Getting Started" was too ambiguous and inclined people to ask "How do I run this program?" questions there occasionally
01:21:56BigBambiyes, I agree
01:22:17 Quit moos (Read error: 110 (Connection timed out))
01:22:46LloreanShould "feature ideas" be moved to development maybe?
01:24:14BigBambinot sure
01:24:36BigBambiTrouble with Feature Ideas is that there is arse all development talk in there normally
01:24:54BigBambiIt is just people saying this'd be cool
01:25:03LloreanIf it were under "Development" maybe that'd be a clue what it's actually _for_ in a way.
01:25:09LloreanBut I also just think the board is a bit imbalanced right now.
01:25:27BigBambiLlorean: You have faith in humanity :)
01:26:06LloreanI also kinda want to put hardware somewhere else
01:26:10LloreanBut there's nowhere to put it.
01:26:14LloreanI'd like to split up general, basically
01:26:36BigBambiYe, it looks too big right now
01:26:52LloreanInto something representing the project itself (webpage, community, etc), something for using the program (most of the existing categories), etc.
01:28:02BigBambiThat's be good
01:29:53BigBambiI think feature ideas could probably move
01:31:48LloreanBigBambi: Okay, how's the new layout?
01:32:02*Llorean thinks this also, slightly, clarifies that Plugins is for discussion of the included ones, not development of them.
01:32:59BigBambiLlorean: I like it
01:34:49 Quit timc (Connection timed out)
01:42:02kugelUnhelpful: yes
01:46:20Unhelpfulkugel: the loading thread generally only does work during scrolling... perhaps you'd best start there. i can't imagine i/o is *slow* on fuze, though...
01:46:33 Quit pixelma (Nick collision from services.)
01:46:35 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
01:46:41 Quit amiconn (Nick collision from services.)
01:46:43 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
01:46:55 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
01:47:03 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
01:48:44 Quit Hillshum ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
01:53:21kugelUnhelpful: *i*? :)
01:54:57Unhelpfulkugel: the thread that loads cache items into memory. it's basically constantly busy while scrolling. the render (main) thread doesn't do any more work at once while scrolling than while not - every screen update == draw the whole screen again.
01:56:00*amiconn still wants to know what's the cause of the missing slide bug in PF on the beast
01:56:10Unhelpfulamiconn: the what?
01:57:13amiconnI've put album .bmps on all my targets which support PF. On the beast, two albums show the ? slide which do have a bmp
01:57:54Unhelpfulamiconn: even when they're centered? :/
01:58:30amiconnOne of the non-working BMPs is even bitwise identical to another one in a different folder, which does work
01:58:59amiconn(it's a two-disc album, and I store two-disc albums in subfolders 'CD1' and 'CD2')
01:59:25Unhelpfulare all of yours simply cover.bmp?
02:00:13rasherDid you ever tell us if it worked in the WPS or not?
02:00:25Unhelpfulweird... i had hoped it might be a funky case issue, but if you're not using <album>.bmp, that seems less likely
02:00:39amiconnAnd I'm using the same directory structure and the same BMP files on all the targets, except the archoses and the c200
02:01:12Unhelpful(to clarify, the "file not found" issue i suspect is somehow case-related - even the rockbox.tar trick doesn't work on my beast if i create the /content dir with the wrong case)
02:01:32 Join Zarggg [0] (
02:01:34rasheramiconn: Does the missing cover work in the WPS?
02:01:45*amiconn still needs to try
02:02:31Unhelpfulamiconn: by the way, have you tried the M3 at all during or since the PF keymap rework? it's never worked in sim, because of remote buttons, so it's basically the only supported target i've not verified...
02:05:07amiconnrasher: Yes, the two missing covers do work in an AA WPS
02:05:34*rasher washes his hands
02:05:41Unhelpfulamiconn: that's bizarre in the extreme, because the same search function is used :/
02:06:01rasherIt must be the loading that fails, then?
02:06:05Unhelpfulperhaps adding a logf on cover not found might be a *start* at debugging it?
02:06:57Unhelpfulrasher: i don't really see how... loading can fail for EOM (pretty much impossible) or for file not found (but then a cover shouldn't have been located?)
02:07:38UnhelpfulEOM is very unlikely, as the scaler only needs the output buffer to be as large as the desired output slide, + 12 rows, and it has the entire plugin buffer
02:07:42rasherYeah, really no idea..
02:11:21 Quit SirFunk_ (Read error: 110 (Connection timed out))
02:11:53Unhelpfulhrm, that's <256KB even for a full-screen load on beast
02:13:35amiconnI've checked the .pfraw files, as the numbering matches the order in PF itself. The .pfraw files are indeed missing, so it's the cache building that fails
02:16:05Lloreansoap: Removed your post, added a bit to mine. Need more category names though.
02:19:09Unhelpfulamiconn: it also splashes on failing to load the bmp, or failing to write the pfraw. so maybe get_albumart_for_index_from_db fails?
02:20:54Lloreansoap: Oh, and updated all the subjects (so that your post became irrelevant)
02:21:19amiconnUnhelpful: Yes, that it what fails
02:21:44*amiconn just added a splash that tells me the index of a failing lookup
02:24:13Unhelpfulso it's either failing to get a track for the album from the DB, which seems unlikely, or the id3 entry is somehow populated differently than when core loads AA, or core's search_albumart_files is somehow failing when PF calls it, but not when the core buffer thread calls it...
02:28:00***Saving seen data "./dancer.seen"
02:28:31 Join AndyIL [0] (n=pasha_in@
02:28:35soapI just moved over the ones which came up on some simple searches. Didn't dig deep.
02:28:44*rasher senses he's not off the hook yet
02:29:33amiconnOne step deeper in the call chain: search_albumart_files() is what fails (like it does for those covers which are actually missing)
02:30:34 Quit Thundercloud (Remote closed the connection)
02:31:21rasherThis does not make sense. Especially since you just use cover.bmp?!
02:31:48amiconnIt's a mystery, but it's also a fact
02:31:59amiconnI know that the core uses the very same function
02:32:09rasherI guess it's not looking in the right dir when looking for cover.bmp?
02:32:59Unhelpfulperhaps the struct mp3entry (somehow) has different contents? how is it filled by core?
02:33:49rasherCan codepage stuff be involved? Does it use non-ascii characters?
02:33:59rasherThe path
02:38:32 Join vonsoo [0] (
02:38:38 Part vonsoo ("Leaving")
02:39:23 Nick fxb is now known as fxb__ (
02:40:53amiconnNo it doesn't. And if it would be a path issue - why does it only happen on the beast, but neither on H340 nor ipod color nor H180 nor X5 nor...
02:41:35amiconnActually, one of the paths does involve non-ascii, but the other doesn't
02:42:56rasherWell it's not like we have any *obvious* guesses left
02:44:07amiconnReally weird issue: The two albums in question do exist twice on my beast - once in mp3 format, and a second time in ape format
02:44:20amiconnGuess where PF looks...
02:44:52amiconnThe ape version is in my test folder - without cover bitmaps...
02:45:21Unhelpfuldoes your test folder have a database.ignore?
02:45:41LloreanAnd the WPS was showing them when played from filetree, rather than database, I assume?
02:46:01rasherLlorean: won't matter, unless amiconn specifically plays the APE
02:46:15amiconnLlorean: yes
02:46:24rasherI'm not sure we can blame this on anything but user error
02:46:28amiconnBut that shows something - if an album exists more than once in the db, PF only checks the first occurence for cover bitmaps
02:46:31Lloreanrasher: that's what I mean though. They showed in the WPS, which would happen if you directly selected the MP3, but I assume the copy visible in the database is probably also the .ape.
02:47:08 Quit linuxstb (Read error: 113 (No route to host))
02:47:09rasherLlorean: won't it appear in the db twice?
02:47:14*Llorean thinks "Duplicate Album Support in multiple formats" isn't really something that's even been attempted.
02:47:15 Quit AndyI (Read error: 110 (Connection timed out))
02:47:23amiconnThe database shows the album once - with all tracks duplicated...
02:47:25Lloreanrasher: I don't know.
02:47:36Unhelpfulamiconn: and PF fetches the first track only from the DB.
02:47:43amiconnI know
02:47:45rasheramiconn: that would be as expected..
02:47:49Unhelpfulif that's the APE...
02:48:10amiconnLlorean: It's not necessarily multiple formats - having the same files in two different folders is enough
02:48:25*rasher suggests keeping albumart in /.rockbox/albumart :)
02:48:38amiconnrasher: I can't
02:48:42*Llorean thinks doing that is the proper solution if you'll have multiple copies of one album.
02:48:54rasheramiconn: Uhm.
02:48:59amiconnSeveral album titles contain characters which are illegal on vfat
02:49:12rasheramiconn: It seems you haven't read the manual
02:49:20amiconnAnd I rather not mangle my tags - it's enough that I had to mangle the folder names
02:49:42amiconnBesides, I don't want the albumart separate from the tree
02:50:10rasherThen this will happen as long as you don't put albumart in both folders..
02:50:38*amiconn will put a database.ignore file insteas
02:51:09Unhelpfulsurely the tags are mangled in memory before generating paths from them?
02:51:28*amiconn once read the manual....
02:51:39amiconnThat was long before album art... even before swcodec
02:51:58amiconnIirc it was the manual for rockbox 2.0
02:53:08rasherBut yes, you can. You chose not to.
02:53:27 Join MethoS [0] (
02:55:50 Quit tvelocity (Remote closed the connection)
02:56:25amiconnRebuilt db and pf cache - now the covers are there...
02:58:45 Quit kushalone ("This computer has gone to sleep. Macs are like babies. They need a lot of sleep. :)")
02:59:55kugelillegal chars are replaced by a "-", and AFAIK rockbox will still recognize the covers
03:00:18rasherkugel: Wrong
03:00:27rasherApparently you haven't read the manual either!
03:00:28Unhelpfulkugel: it's an underscore, except for ", which is replaced with '
03:00:45rasher(within the last month)
03:01:12*amiconn wonders whether it would make sense to check more tracks entries from the db if the first one doesn't yield a cover bitmap
03:01:37amiconnMaybe limited to a track index of 1 (that's part of struct mp3entry id3)
03:02:18Unhelpfulthey'll be sorted by track index, right?
03:02:39 Join webguest63 [0] (n=4b8ffbfa@gateway/web/cgi-irc/
03:03:11rasherSeems quite unlikely, but it's just PF, so it can't hurt if someone bothers I guess
03:03:37kugelrasher: "within the last month"?
03:04:06rasherkugel: The album art chapter was added to the manual a month ago or so. It's been on the AlbumArt page for a
03:04:23kugelah yea
03:04:43kugelI was fairly sure about the "-" though, I'm wondering where I picked that up :/
03:09:54 Join swimmer [0] (
03:14:25 Quit Lss (Read error: 104 (Connection reset by peer))
03:15:15 Join shodanX [0] (
03:16:36 Quit shodanX__ (Read error: 60 (Operation timed out))
03:17:36 Quit webguest63 ("CGI:IRC")
03:24:36kugelAny idea why putting the lcd framebuffer into iram let's the binsize raise so much?
03:26:19kkurbjunkugel, are you using the (NOLOAD) setting, and is it being put in the .ibss?
03:26:34kugeluhm, good question
03:28:16kugelkkurbjun: no, not in ibss. let me try that
03:28:54kugelour ibss uses no load, iram doesn't
03:29:22kkurbjunyeah, the bss section is for uninitialized values
03:29:38kkurbjunso you don't have to store a copy in the image
03:29:55kkurbjunwell bss and ibss
03:30:07kugelindeed, that did it
03:30:21kugelthank you!
03:30:27kkurbjunno problem
03:31:10kkurbjunkugel, about the mmu problems you were having.. I was running into a large slowdown when my screen routines in the m:robe were not staying within the screen boundary
03:31:26kkurbjunso I was accidentally overwriting the ttb with the screen updates
03:31:50kugelwe don't have the framebuffer in an extra section
03:32:18kkurbjunyeah, but how do you calculate the FRAME address?
03:32:21kugel(i.e. the ttb shouldn't be touched by it)
03:32:36kkurbjunit was still a problem without giving it a section
03:33:19kkurbjunI guess now that it's in iram it should be safer anyway
03:33:32kkurbjunis your lcd not memory mapped?
03:33:46kkurbjunoh, gotcha
03:33:52kkurbjunnevermind on that then
03:35:12kugelkkurbjun: there's no caching at all in that patch, that may explain the slowness
03:35:41 Join z35_ [0] (
03:35:41*kugel is wondering if the icache is still enabled properly in that patch.
03:35:52kugelI had similar slowness when I deactivated the icache
03:47:06 Quit z35 (Read error: 110 (Connection timed out))
03:50:02 Nick z35_ is now known as z35 (
03:51:00 Join Strife89 [0] (n=nds@
03:51:28 Quit Strife89 (Client Quit)
03:53:13 Quit davidfg4 ()
03:54:00 Join dreamlayers [0] (
03:55:19 Join Strife89 [0] (n=nds@
03:56:47 Join froggyman [0] (n=47ba40e2@gateway/web/cgi-irc/
03:57:51dreamlayersI'm wondering what's the point of using read-modify-write operations (eg. &=) to set or clear PP502x GPIO bits when there are bitwise operations.
03:58:45 Join Dhraakellian [0] (
04:01:30froggymancould i get the wiki wrire oermissions from someone?
04:03:51kkurbjunfroggyman: what's your registered name?
04:07:10kkurbjunI think I added you, but I'm not positive
04:09:04kkurbjunIf you still can't write ask back and someone who is more familiar with the process might be around, no problem
04:09:37froggymanok i will
04:15:32 Quit intrados_ (Remote closed the connection)
04:18:02froggymanit appears that i do have the write permissions now
04:24:58dreamlayersWhat's the status of FS #8806 - MikMod MOD, S3M, IT, XM player? It seems like such a great addition to Rockbox.
04:25:13 Quit DataGhost (Nick collision from services.)
04:25:21 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
04:26:21kkurbjunLlorean: I tried adding froggyman to the wiki users group, but it doesn't seem like he has permissions to write, is there something else that needs to be done?
04:28:02***Saving seen data "./dancer.seen"
04:29:11froggymanare you sure, because i just checked the main page and it says that i edited something
04:29:26kkurbjun:), I misread what you wrote
04:29:35kkurbjunnevermind, Llorean
04:29:48kkurbjungreat, glad to hear it works
04:38:09Strife89dreamlayers: I agree; good question.
04:38:39 Join Nordheim [0] (
04:39:12*kugel has a little hackish way to make the scrollwheel like perfect on the fuze
04:39:20NordheimI can't seem to get the Rockbox automated install to load correctly
04:39:42Nordheimit autodetects my device fine, but says the mount point is not writeable, whether run in root or not
04:40:26Nordheimdevice [a Sansa E260] is in MSC mode already
04:44:22Nordheimshould it be pointed to the mount point or actual device ?
04:46:37 Join hillshum [0] (n=hillshum@unaffiliated/hillshum)
04:51:40 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:05:38kugelhuh, with this hack, the scrollwheel is almost better than on e200v1
05:11:22 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
05:18:26 Quit hillshum ("Leaving")
05:18:55re-peatwhat does mount say about it?
05:19:50 Quit franky (Read error: 113 (No route to host))
05:21:14 Quit l403 (Read error: 113 (No route to host))
05:22:28 Quit froggyman ("CGI:IRC (EOF)")
05:22:50 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
05:24:43Nordheimyeh, it is showing as read-only
05:24:54NordheimI have since gotten rockbox installed, but now I can't add/remove any files
05:26:01saratogaon linux that usually means youv'e corrupted the file system
05:26:23Nordheimyes, it does seem to be that
05:26:29Nordheimthat's the main reason I'm trying rockbox
05:26:55 Join froggyman [0] (n=47ba40e2@gateway/web/cgi-irc/
05:26:57NordheimI start getting lots of 'bad track' errors in the default
05:33:30 Join SirFunk [0] (
05:34:46 Quit Strife89 ("ClIRC - IRC client for Nintendo DS")
05:38:01Nordheimpost-format seems to be working fine
05:39:48 Quit kugel (Read error: 110 (Connection timed out))
05:40:31 Quit MethoS (Remote closed the connection)
05:48:51 Join intrados_ [0] (
05:57:37 Quit perrikwp (" ajax IRC Client")
05:59:39 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
06:06:17 Quit froggyman ("CGI:IRC (EOF)")
06:28:03***Saving seen data "./dancer.seen"
06:37:20 Quit dreamlayers ("Goodnight")
06:45:24 Join Adam917 [0] (
06:46:10 Part Adam917
07:02:15 Join DJ_Igloo [0] (n=4b24625a@gateway/web/cgi-irc/
07:03:15 Quit DJ_Igloo (Client Quit)
07:10:51 Join lightbul1jim [0] (
07:16:21 Join jeffdameth [0] (
07:22:03 Quit Zoxc ()
07:24:00 Join merbanan [0] (
07:25:41 Quit hittudiv (Remote closed the connection)
07:26:56 Quit lightbulbjim (Read error: 110 (Connection timed out))
07:33:55 Quit jeffdameth1 (Read error: 110 (Connection timed out))
07:41:05 Quit timc`` (Read error: 104 (Connection reset by peer))
07:41:14 Join timc`` [0] (n=aoeu@
07:46:02 Join bs66_1 [0] (
07:59:42 Quit Nordheim ("Get me closer, I want to use my sword.")
08:03:08 Quit bs66_ (Read error: 110 (Connection timed out))
08:09:37 Quit rvvs89 (Read error: 113 (No route to host))
08:14:08 Join stoffel [0] (
08:21:33 Join Minthe [0] (
08:28:06***Saving seen data "./dancer.seen"
08:30:01 Quit saratoga ("CGI:IRC (EOF)")
08:31:12 Quit z35 ("Leaving")
08:37:15 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
08:38:05 Join franky [0] (
08:39:04 Part BHSPitLappy ("Ex-Chat")
08:55:03 Join Rob2222 [0] (
08:55:11 Quit _lifeless (Read error: 110 (Connection timed out))
08:59:52 Quit rwcr (Read error: 54 (Connection reset by peer))
09:03:04 Join flydutch [0] (
09:11:18 Join rwcr [0] (
09:12:22 Quit Rob2223 (Read error: 110 (Connection timed out))
09:24:35 Join _lifeless [0] (n=lifeless@
09:29:26 Join SuperPerroFlashi [0] (
09:31:34 Quit SuperPerroFlashi (Client Quit)
09:35:17 Join l403 [0] (n=l@
09:37:45 Nick fxb__ is now known as fxb (
09:40:26 Join bertrik [0] (
09:47:34 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
09:49:51 Nick fxb is now known as fxb__ (
10:23:19 Quit stoffel (Read error: 113 (No route to host))
10:26:01 Join arez-l [0] (n=4fea9651@gateway/web/cgi-irc/
10:28:09***Saving seen data "./dancer.seen"
10:30:19 Join wangxiang [0] (n=wangxian@
10:30:20 Join miepchen^schlaf [0] (
10:36:46amiconnkkurbjun: Did you magically enlarge the m:robe remote lcd?
10:40:40 Join rvvs89 [0] (
10:43:49arez-lcould anyone help me out compiling the bootloader for a sansa e200v2 ? finished tools/configure and also tried make but i cannot find the created file
10:45:25 Join hittudiv [0] (n=hittudiv@
10:45:38bertrikthe bootloader image is called bootloader-clip.sansa
10:46:10arez-lits not there
10:46:42arez-lso doing tools/configure and then using make is right?
10:47:28bertrikno, you should build it in a separate build directory
10:49:38arez-liam quite new to compiling.. is there a quick guide for this? the wiki requires some compiling knowledge as it seems
10:51:43 Quit rvvs89 (Remote closed the connection)
10:51:46arez-lor could you give me some short instructions?
10:58:34bertrikthere are several guides on the wiki, but I'm not really familiar with them
10:58:57arez-li just followed one but it only says that i need to create the bootloader with tools/configure
10:59:43arez-lhow to tell it to use a seperate build folder?
11:01:35bertrikwhat I usually do (from the main rockbox source folder): mkdir build-clip-boot; cd build-clip-boot; ../tools/configure; (select 55-clip, B for bootloader); make
11:01:51 Quit merbanan ("Leaving")
11:02:01arez-lgonna try that now, thank you
11:03:18bertrikactually, this is just the easy first part of installing rockbox on an ams sansa ...
11:03:24arez-lgetting some errors but idk what they mean
11:05:40bertrikbuilding the bootloader should be done in a different directory than building the main rockbox software, your build directory is simply called build so I think you're mixing the two. A make clean and another make may help if you get weird compilation errors.
11:06:21 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
11:06:33arez-lso i sgould use a build bootloader folder for this and try again after using make clean
11:08:21 Join ender` [0] (
11:09:50bertrikyes, use something like 'build' for the main firmware and something like 'build-boot' for the bootloader. A make clean simply cleans up old build results that may confuse the build process.
11:10:49arez-lgot it now :)
11:11:23 Join Adam917 [0] (
11:12:33 Quit arez-l ("CGI:IRC")
11:13:32Adam917Anyone here know if the instructions found on the 'BigDisk' page apply to the iriver units as well? As in would I need to compile a multi-partition variant of Rockbox and have to manually copy files to a second partition after 128 GiB? What about USB support?
11:22:45 Join arez-l [0] (n=4fea9651@gateway/web/cgi-irc/
11:23:12arez-lhow do i solve "command not found" when i try to use mkamsboot for patching the firmware?
11:23:55 Join rvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
11:24:20bertrik'make' sure that the command is present? what command could not be found?
11:25:33arez-lalready tried "make" in the mkamsboot folder
11:27:32arez-lnvm ; worked after make clean and another make
11:27:38bertrikyou should type ./mkamsboot
11:28:53arez-lhow do i know if the bootloader works ? i dont have compiled a rb build so far
11:31:04bertrikit briefly shows a rockbox logo (and probably an error about not finding the .rockbox directory if you haven't installed the main firmware yet)
11:32:13 Join MT [0] (n=chatzill@
11:33:19bertrikIIRC, holding the home button during boot or booting by plugging in USB from the power-off state makes it boot into the original firmware again. In case of hangs, hold the power button for up to 20 seconds.
11:34:12arez-lthat might be useful :)
11:34:33arez-lgotta reboot the vm here ; brb
11:35:03 Quit arez-l ("CGI:IRC (EOF)")
11:35:15 Join Seed [0] (
11:38:01 Join {phoenix} [0] (
11:43:49 Join einhirn [0] (
11:46:14 Join stoffel [0] (
11:57:27 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
12:01:41 Quit Minthe ("Leaving...")
12:13:53 Quit bmbl ("Woah!")
12:17:37 Quit swimmer (Client Quit)
12:19:24 Join moos [0] (i=mustapha@rockbox/staff/moos)
12:28:12***Saving seen data "./dancer.seen"
12:35:19 Join pyro_maniac [0] (
12:38:02 Join MethoS [0] (
12:43:17 Nick fxb__ is now known as fxb (
12:50:23 Join tvelocity [0] (
12:54:40 Join Thundercloud [0] (n=thunderc@
12:57:39 Quit kadoban (Remote closed the connection)
13:06:43 Quit Thundercloud (Read error: 104 (Connection reset by peer))
13:20:05 Join Ubuntuxer [0] (
13:29:37 Join midijunkie [0] (
13:41:22 Quit moos ("Rockbox rules the DAP world")
13:46:09 Quit Ubuntuxer ("Leaving.")
14:06:37 Join intrados [0] (
14:08:06 Join gregsparc [0] (n=chatzill@
14:16:54 Quit Tomers ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
14:22:35 Quit intrados_ (Connection timed out)
14:28:16***Saving seen data "./dancer.seen"
14:34:51*Adam917 wishes Rockbox were on phones as their audio player
14:36:57 Quit Seed ("cu, Andre")
14:37:46pixelmaI just noticed that the volume level shown in an Ondio sim is off compared to the target (ranging from -40dB to +60dB, whereas it's -100dB to +12dB in reality)
14:38:50pixelmahwcodec sims don't actually play music but I thought volume range could at least show the same
14:41:48 Quit midijunkie (Read error: 104 (Connection reset by peer))
14:50:51 Quit intrados (Remote closed the connection)
14:52:37 Join intrados [0] (
14:53:29 Quit intrados (SendQ exceeded)
14:54:40 Join intrados [0] (
14:54:54 Join AndyI [0] (i=AndyI@
14:57:00 Join tomers [0] (
14:58:22tomersHello, I've added USB_SERIAL to my iPod. What device should I use to view serial traffic? I have /dev/usbdev1.1_ep00, /dev/usbdev1.1_ep81, /dev/usbdev2.1_ep00, /dev/usbdev2.1_ep81. Is it one of those?
14:59:49gevaertshm, that's weird... You should have /dev/ttyUSB0. Did you load usb_serial as described on ?
15:00:38 Join kugel [0] (n=kugel@rockbox/developer/kugel)
15:03:12tomersNo, I did not consulted that page. Thanks a million. I added it in frimware/export/config-XXX.cfg... I'll fix it, and update in case it doesn't work.
15:04:26gevaertsActually, the usb_serial implementation isn't very complete yet. It's good enough for linux to talk to, but it doesn't fully implement the CDC spec so windows doesn't like it at all.
15:05:21gevaertsAlso, thanks for spotting my unintentional advanced C feature knowledge test ;)
15:06:05 Quit feisar (Read error: 104 (Connection reset by peer))
15:06:37 Join feisar [0] (
15:09:52tomersThanks! I'll update the cosmetic patch to fit the current code convention of these files (It's not the important currently, and it takes too much time to do...)
15:14:49 Quit AndyIL (Read error: 113 (No route to host))
15:16:11gevaertsDon't worry too much about cosmetics :) Working right wins against looking right every time!
15:21:48*gevaerts formally denies allegations that he broke the build
15:22:32gevaertsWhy wasn't r20684 autobuilt?
15:25:08pixelmasometimes builds aren't picked up for unknow reasons
15:25:17pixelmaunknown too
15:26:40 Join xwang [0] (n=wangxian@
15:27:10 Quit xwang (Client Quit)
15:28:03 Join sko [0] (
15:28:52 Quit tomers (Remote closed the connection)
15:29:43 Part sko
15:29:47pixelmaseeing that there are also main display this size - could the theme site have subpages for 128x96x2 (Iaudio M3/remotes) and/or 128x64x1 (Clip, m200, other new ports, Iriver remotes) displays?
15:34:43 Quit MethoS (Read error: 104 (Connection reset by peer))
15:34:43 Join MethoS- [0] (
15:35:04gevaertsAny reason why the Official Test Builds forum isn't just restricted to only allow staff to start new threads?
15:35:08gevaertsOr is it?
15:35:58 Join tomers [0] (
15:36:30 Join SirFunk_ [0] (
15:43:30 Quit ufoman_ (Remote closed the connection)
15:43:42 Quit wangxiang (Read error: 110 (Connection timed out))
15:49:08 Join shodanX_ [0] (
15:49:22 Quit timc`` (Read error: 104 (Connection reset by peer))
15:49:41 Join timc [0] (n=aoeu@
15:51:17 Quit SirFunk (Read error: 110 (Connection timed out))
15:52:02 Join motionman95 [0] (n=459792ac@gateway/web/cgi-irc/
15:52:54 Quit motionman95 (Client Quit)
15:54:14 Join motionman95 [0] (i=459792ac@gateway/web/ajax/
15:54:46tomersI have some more USB questions. I can in wiki (/PortalPlayerUsb#logf_over_USB_Serial) see that USB_SERIAL should be defined in usb_core.h - it is actually there, commented out. But I also see it is in config-iaudio7.h (commented out also). is this some kind of redundancy, or should I set this config both in usb_core.h and the config file (config-ipodcolor.h)? If I should set it in config-ipodcolor
15:54:48tomers.h, can I just rebuild it using 'make', or should I remove and re-create the build directory?
15:55:50motionman95Could someone explain to me how to use the read_line() function it's perplexing me, and is not used often enough in plugins for me to understand exactly how I should use it.
15:56:34rashermotionman95: docs/PLUGIN_API hopefully explains a bit
15:56:47 Join Ridayah_ [0] (
15:57:00motionman95Okay, I'll read it and be right back...
15:57:53gevaertstomers: I originally put it in usb_core.h, and described that on the wiki. I guess vitja either didn't see that or disagreed with it and put it in config-iaudio7.h. Right now I'm not sure which is best :)
15:57:53rasherThat file isn't updated much, but many functions are there
15:58:06 Quit Ridayah_ (Client Quit)
15:58:35 Join wangxiang_ [0] (n=wangxian@
15:59:30gevaertstomers: I don't know if the build system will pick it up correctly either way, so probably starting with a clean build dir is safest
15:59:44motionman95@rasher: Okay, so based on what I read, read_line() works like this: the first time it's called, it will read up to the first LF, the second time, it will read up to the second LF, and so on, until it returns -1?
16:00:10rashermotionman95: Pretty much
16:01:02 Quit Ridayah (Read error: 110 (Connection timed out))
16:01:18motionman95Thanks. Now, instead of bugging the people on the Rockbox forums, I guess I'll just post here.
16:02:01 Quit einhirn (Read error: 104 (Connection reset by peer))
16:02:03 Join Ridayah [0] (
16:03:16tomersgevaerts: OK. I'll start with a clean build. But can I just replace the .ipod file, and reboot the DAP?
16:03:51gevaertstomers: either that, or put the new .ipod file somewhere else and "play" it
16:04:40tomersgevaerts: Can I replace the the existing .ipod file while rockbox is connected (in charging mode) ?
16:06:12tomersgevaerts: It seems the answer is yes :-)
16:06:28gevaertsOnce the bootloader (or rolo) has loaded it, nothing touches that file anymore
16:06:38gevaertsExcept you of course :)
16:07:49tomersgevaerts: Damn. The iPod battery is a bit drained. It doesn't charges well, and I have to wait too long before the bootloader loads rockbox (several minutes) each time
16:08:19gevaertsTry charging in the OF (or in disk mode). That may work better
16:12:36 Join fdinel [0] (
16:19:35motionman95Peace out. :)
16:22:30 Part motionman95
16:24:46 Join motionman95 [0] (n=459792ac@gateway/web/cgi-irc/
16:26:04 Quit motionman95 (Client Quit)
16:28:18***Saving seen data "./dancer.seen"
16:38:21 Quit pyro_maniac (Read error: 104 (Connection reset by peer))
16:41:20 Join z35 [0] (
16:46:44 Quit DarkSpectrum (Read error: 104 (Connection reset by peer))
16:47:23 Quit DZX (Read error: 104 (Connection reset by peer))
16:47:44 Join wincent [0] (
16:53:01 Join Pepsi1x1 [0] (
16:54:36 Quit kugel (Read error: 110 (Connection timed out))
16:57:14 Join kugel [0] (
16:57:23Pepsi1x1hi, i have an ipod mini generation 2, and I would like to install rockbox, although when i am trying to use the rockbox utility, I have an error first from autodetect saying it is an incompatible version, but it is in the list of players. Also if I try to manually select it, when I click complete install it errors straight away saying Ipod not detected. I have looked through the manual, and faq's and
16:57:41Pepsi1x1a google search returned only poor youtube videos
16:57:53gevaertsAre you sure it's a mini and not a nano?
16:57:59*kugel has a little make install hack for normal builds
16:58:10Pepsi1x1yes the model number is ma447
16:58:21Pepsi1x1let me double check the numbers
16:58:33tomersI added info on how to enable logf (using advanced build) in Wiki page PortalPlayerUsb
16:58:57Pepsi1x1hah typos
16:59:34gevaertsgoogle tells me that that's a nano
16:59:48Pepsi1x1really it told me it was a mini
17:00:05gevaertsHave a look at to make sure
17:00:39Pepsi1x1oh dear
17:00:44Pepsi1x1yes its a nano
17:00:55Pepsi1x1how foolish was that
17:01:20Pepsi1x1nano 2nd gen is unsupported then, if im not mistaken?
17:01:35gevaertsunfortunately, yes
17:01:43Pepsi1x1ok thanks for your help
17:04:26 Join Thundercloud [0] (n=thunderc@
17:05:35 Quit miepchen^schlaf ()
17:06:33 Quit {phoenix} (Remote closed the connection)
17:21:00Pepsi1x1so is there any good alternative to rockbox which is compatible with the 2nd gen nano?
17:21:07 Quit bs66_1 (Read error: 60 (Operation timed out))
17:22:09 Join bs66_ [0] (
17:23:47 Quit advcomp2019 (Read error: 113 (No route to host))
17:34:08 Join AndyIL [0] (n=pasha_in@
17:38:35 Join SirFunk__ [0] (
17:39:57 Quit tomers (Remote closed the connection)
17:40:04 Join motionman95 [0] (i=459792ac@gateway/web/ajax/
17:40:05 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
17:41:20motionman95Is there a special room to chat in for WPS help?
17:41:50 Join tomers [0] (
17:42:21kugelis this too hackish?
17:44:01kugelthat's make install for a normal build to install directly (well, indirectly using tar and extracting then) on the target (path specified by PREFIX)
17:44:34 Quit AndyI (Read error: 110 (Connection timed out))
17:45:18 Join CaptainKewl [0] (
17:52:29 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
17:53:14 Quit SirFunk_ (Read error: 110 (Connection timed out))
17:53:24 Join Lss [0] (
17:54:38 Quit feisar (Read error: 104 (Connection reset by peer))
17:55:38 Join xt3 [0] (
17:55:42 Join z35_ [0] (
17:57:21xt3can anyone help me with firmware for an onda vx747. i compiled it and tried to put it on the player but whenever i tried it the player stuck with a message "ata-18 panic"
18:00:40 Quit z35_ (Read error: 54 (Connection reset by peer))
18:02:57 Join merbanan [0] (
18:03:07 Quit blithe ("Lost terminal")
18:03:16 Quit gevaerts (Nick collision from services.)
18:03:17 Join blithe [0] (
18:03:24 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
18:03:29 Join z35_ [0] (
18:05:09 Quit z35 (Read error: 110 (Connection timed out))
18:06:15kkurbjunamiconn, I did that because rockbox was freezing, I think something is overflowing the lcd
18:06:32kkurbjunso I gave it more room to keep it from crashing for now
18:07:56kkurbjunthe remote LCD on the m:robe is much smaller than any of the others 79x16, so I was going to try searching for anything that had hard-coded or calculated offsets/width/heights that could cause an overflow
18:11:57kkurbjunamiconn, if you have any ideas on where to start looking it would be much appreciated. Otherwise I might talk to cat and see if he can connect his JTAG setup so I can do some remote debugging again (or solder my own)
18:14:26 Part xt3
18:23:10 Join dreamlayers [0] (
18:28:21***Saving seen data "./dancer.seen"
18:32:33 Nick l403 is now known as invisible_pie (n=l@
18:40:55 Join stealthboy [0] (i=459792ac@gateway/web/ajax/
18:42:30stealthboyHow dificult would X and Y be to implement for drawing pictures to the screen? Say, instead of having to constantly redraw to the LCD, simply assign what was drawn to the screen a new X and Y?
18:43:57kkurbjunstealthboy: most targets implement lcd_update_rect
18:44:12kkurbjunwhere you give it an x, y, width, height
18:44:25kkurbjunand it only copies the section that needs to be updated
18:45:19 Join kps00000 [0] (
18:46:21 Quit wangxiang_ ()
18:47:43kugeleh, is broken, somehow
18:49:53 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
18:52:21 Join robin0800 [0] (
18:52:23dreamlayerskugel: works for me when building 5G firmware zip, and it was last modified 2 months ago. Which target causes it to break?
18:52:50kkurbjunamiconn, looks like I can get it to freez in the simulator too
18:53:37 Quit timc (Read error: 104 (Connection reset by peer))
18:53:48 Join timc [0] (n=aoeu@
18:54:27 Quit tomers (Read error: 104 (Connection reset by peer))
18:57:38 Join tomers [0] (
18:57:48 Quit tomers (Client Quit)
19:02:29 Quit flydutch ("/* empty */")
19:06:57 Quit robin0800 (Remote closed the connection)
19:07:42kugeldreamlayers: nevermind, I blame perl
19:08:34kugelusing @ARGV for essential parameter seems fragile
19:12:23 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:12:56 Nick z35_ is now known as z35 (
19:14:05dreamlayersHow concerned should I be with code size? When I was working on a patch with some fixes and improvements of 5G iPod LCD sleep, I spent too much time wondering about this. For example, when waiting for the LCD to finish initializing, do I use a wakeup struct, or do I go for smaller code size and use a polling loop which yields?
19:15:12 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
19:17:19 Join jeffdameth1 [0] (
19:20:36 Part kps00000
19:20:40 Join kps00000 [0] (
19:23:45kugeldreamlayers: well, code size isn't as much of an issue on a target with 32/64MB RAM. but every increase slightly decreases the battery runtime
19:24:06kugel(particularly on HDD targets)
19:26:40 Quit hittudiv (Remote closed the connection)
19:32:38 Quit jeffdameth (Read error: 110 (Connection timed out))
19:47:05 Quit stealthboy (" ajax IRC Client")
19:50:17 Join paulk [0] (
19:50:42 Quit stoffel ("leaving")
19:51:55 Join motionman [0] (i=459792ac@gateway/web/ajax/
19:54:50paulkHello ! When i connect my sansa e250 to my computer, the original firmware is launched by rockbox. How to connect the player without using the original firmware please ???
19:57:11motionmanpaulk: You can do this by using the current build, or applying the bootloaders found in the flyspray.
19:57:31paulkI'm using the current build
19:58:30saratogawhat revision number?
19:59:00paulkI've got rockbox 3.2
19:59:42saratogathats not the current build
19:59:48saratogaupdate and you'll get USB
19:59:56paulkwhere can i found a bootloader able to do it ?
20:00:24saratogayou don't need a bootloader, just update rockbox
20:00:52paulk? I've got the lastest version
20:01:14gevaertsif you want the bootloader to load rockbox on plugin, you need a new bootloader as well
20:01:24gevaertspaulk: 3.2 is the latest *release*
20:01:36paulkok, it's the lastest build
20:01:46paulkit's not the lastest build *
20:02:23paulkok, where do i put the pp5022.mi4 file ?
20:02:54 Join Ubuntuxer [0] (
20:03:38kugelpaulk: use it with sansapatcher
20:03:49kugeldont put the file directly anywhere on your dap
20:04:03kugel(unless you want to screw up the OF)
20:04:50paulkI've done it with sansapatcher I'll do a test
20:05:40kugelwhat command did you use?
20:05:42paulkit does not work...
20:05:47paulk./sansapatcher -a firmware.mi4
20:05:52paulkas root
20:06:16paulkis it right ?
20:06:20kugelnot sure
20:06:23saratogai think you should just wait for an official bootloader
20:06:29saratogayou don't need to install this thing to get USB
20:06:30kugel./sansapatcher −−help should tell
20:07:07saratogaand that didn't work because you don't have a fimrware.mi4 file
20:07:22paulkit's called PP5022.mi4
20:07:30paulkis it important ?
20:07:48saratogaits the bootloader you're trying to install
20:07:55gevaertsreally, if you ask that sort of questions, stay with the stable build...
20:08:16kugelpaulk: what exactly did you mean by "does not work" btw? the bootloader doesn't feature usb as is. It just doesn't boot the OF if you insert the cable in the OFF state
20:08:16saratogaso its not important in the sense that what you're doing doesn't matter, but its important in the sense that if you do the wrong thing now you'll break your player
20:08:59kugelsaratoga: can you try FS #10120?
20:09:03paulkyes, but it boot the OF
20:09:22kugelthe version should tell you whether the update worked
20:09:52dreamlayerspaulk: If you don't install the bootloader and you want to use Rockbox USB, just make sure that you turn on your e250 before you plug in USB.
20:10:18paulkI do
20:10:31saratogadid you update your build like i told you to 10 minutes ago?
20:10:42saratogabecause that step is important
20:11:08paulkoh no !
20:11:16paulkwhere can i get the lastest build ?
20:11:26saratogathe front page of this website
20:12:37saratogakugel: i'll take a look at it when i get my fuze from work this afternoon
20:12:49saratogabut that approach seems sensible
20:13:41 Join DarkSpectrum [0] (
20:13:54 Join Traveler [0] (
20:14:26saratogahow difficult is it going to be to get the AMS sansas to work with rbutil?
20:14:41 Quit Ubuntuxer ("Leaving.")
20:14:54 Quit timc (Read error: 104 (Connection reset by peer))
20:14:55 Join timc`` [0] (n=aoeu@
20:14:57 Join ultrasonic [0] (n=manohar@
20:15:07gevaertssaratoga: they seem to need approximately the same steps as h100 and h300
20:15:51saratogawget an OF file, run the merge program on it, and then copy to the player?
20:16:09 Quit ultrasonic (SendQ exceeded)
20:16:12saratogaso relatively easy then
20:16:22gevaertsAlthough I think rbutil asks you to get the OF file
20:16:38 Quit kugel (Nick collision from services.)
20:16:43 Join ultrasonic [0] (n=manohar@
20:16:43 Join kugel [0] (n=kugel@rockbox/developer/kugel)
20:16:45saratogasandisk provides http links to the raw OF image, so downloading it should be easy enough
20:17:15kugelgevaerts: what will happen about the version then?
20:17:21kugelI mean the OF version
20:17:35kugeldo we pick a random OF version and always use that or do we try to keep the version installed?
20:17:39saratogathey'll get whatever one is hard coded unless we put an option or they use the manual install
20:17:51 Quit ultrasonic (SendQ exceeded)
20:18:04saratogathe ams patcher program has to be updated for each OF version anyway
20:18:05kugelmaybe there's a way to read the installed OF version
20:18:24 Join ultrasonic [0] (n=manohar@
20:18:25paulkyeah ! It works !
20:18:31kugelwhich reminds me, I wanted to update mksamsboot for 1.02.26
20:18:33gevaertskugel: IIRC rbutil just asks you to tell it where you put the hex file for irivers, so it uses whatever version you give it
20:18:41bertrikcan't we just put on a pre-patched firmware?
20:19:00gevaertsWe can't distribute the OF
20:19:13kugelwe can't?
20:19:19kugelBagder does that
20:19:30 Quit ultrasonic (SendQ exceeded)
20:20:03 Join ultrasonic [0] (n=manohar@
20:20:15kugelsaratoga: Well, the approach might seem sensible, but I nevertheless feel that it's a bit hackish :/
20:20:25 Quit saratoga ("CGI:IRC (EOF)")
20:20:28 Join saratoga2 [0] (n=41becb3b@gateway/web/cgi-irc/
20:20:30 Quit paulk ("Ex-Chat")
20:20:44kugelbut it apparently just works, even nicely
20:20:50saratoga2polling a wheel is a big hack but unless we find a way to do interrupts we have no choice
20:21:09 Quit ultrasonic (SendQ exceeded)
20:21:43 Join ultrasonic [0] (n=manohar@
20:21:45kugelthe button driver is full of hacks anyway :p
20:21:47saratoga2i'm hoping that if we ever get the AS353x datasheet, it will have a way to do dbop interrupts we can use but i'm not counting on it
20:22:33kugelwell, dbop has interrupts. but being rather a bus than a IO thing, it only has interrupts for FIFO etc
20:22:48saratoga2yeah i vaguely remember that
20:22:51 Quit ultrasonic (SendQ exceeded)
20:23:01kugelI hoped we can get interrupts on GPIOB, then switch to DBOP and read, but that doesn't seem possible
20:23:24 Join ultrasonic [0] (n=manohar@
20:23:27saratoga2the physical pins aren't the same IIRC, so its probably not
20:23:37*kugel managed to read power by using an interrupt, but no success for the wheel
20:23:55kugelnow we read power on dbop
20:23:58saratoga2haha good thing sandisk let us get power on interrupts
20:24:34 Quit ultrasonic (SendQ exceeded)
20:24:42kugelbut actually, sandisk seems to do the same
20:24:57saratoga2are there any ams developers who know rbutil or even c++?
20:25:07 Join ultrasonic [0] (n=manohar@
20:25:08kugellooking at the disassembly, it seems to call lcd_button_support (to write the red pixel) too.
20:25:36saratoga2red pixel?
20:26:04kugelwe need to write a red pixel into the dbop fifo in order to read the buttons correctly
20:26:13 Quit ultrasonic (SendQ exceeded)
20:26:45 Join ultrasonic [0] (n=manohar@
20:27:03kugelusing a output bus for buttons is fundamentally flawed anyway. thank you sandisk
20:27:31kugelbut, well, there's only 32 gpio lines, which is just too few, they probably didn't have a choice :(
20:27:43kugelPP has like 100?
20:27:53 Quit ultrasonic (SendQ exceeded)
20:28:25***Saving seen data "./dancer.seen"
20:28:26 Join ultrasonic [0] (n=manohar@
20:28:26saratoga2still you'd think they'd put the hold button or something like that on DBOP, not the most time critical buttons
20:29:35dreamlayerskugel: It seems like PP allows most pins to be used as GPIO, including ones assigned to other functions like IDE. So for example a device which doesn't use IDE can use some those pins for GPIO, but if you want to use IDE, they have to be used for IDE.
20:29:36 Quit ultrasonic (SendQ exceeded)
20:30:09 Join ultrasonic [0] (n=manohar@
20:31:15 Quit ultrasonic (SendQ exceeded)
20:31:22kugeldreamlayers: that's on the as3525 too
20:31:39saratoga2AMS sort of does that, where some DBOP pins are GPIO [iirc] but they have just 32 GPIO logical pins, and more physical pins so some are unmapped
20:31:49 Join ultrasonic [0] (n=manohar@
20:31:54kugellike GPIOD is almost exclusively used for xpd (SD interface)
20:32:05saratoga2which is kind of baffling, physical pins are expensive, logical ones are not
20:32:13kugelWe basically cannot control the buttonlight correctly, because we cannot use it as GPIO
20:32:35Travelerdid the repository get rid of the text viewer?
20:32:57 Quit ultrasonic (SendQ exceeded)
20:33:15 Join midijunkie [0] (
20:33:28 Join ultrasonic [0] (n=manohar@
20:34:38 Quit ultrasonic (SendQ exceeded)
20:35:12 Join ultrasonic [0] (n=manohar@
20:36:20 Quit ultrasonic (SendQ exceeded)
20:36:55 Join ultrasonic [0] (n=manohar@
20:37:10 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
20:37:21 Quit linuxstb (Nick collision from services.)
20:37:32 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
20:38:04linuxstbTraveler: apps/plugins/viewer.c ?
20:38:05 Quit ultrasonic (SendQ exceeded)
20:38:19kugelis the for CodecPerformanceComparison available anywhere?
20:38:41 Join ultrasonic [0] (n=manohar@
20:39:07 Join intrados_ [0] (
20:39:38Travelerah, thanks; was looking for text_viewer.c
20:39:46 Quit Traveler ("Java user signed off")
20:40:17Dhraakellianso is the Fuze (and AMS sansa general) port approaching usability now, if not efficiency?
20:40:45franky./rs moni
20:41:04frankyoops, sorry.
20:44:50 Join DarkSpectrum- [0] (
20:46:43 Quit DarkSpectrum (Read error: 104 (Connection reset by peer))
20:47:39*dreamlayers found some non-critical bugs in flash-recorderv2-v3 BootBox
20:48:12kugelsaratoga2: how did you test mp3 performance?
20:48:39kugeljust do "open with" on a sufficiently small file?
20:49:13 Nick invisible_pie is now known as l403 (n=l@
20:50:01 Quit intrados (Read error: 110 (Connection timed out))
20:52:25kugelah, 4,1MB is too big already :/
20:55:18 Quit CaptainKewl (Read error: 110 (Connection timed out))
21:01:48 Quit jhMikeS (Nick collision from services.)
21:01:53 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
21:02:11saratoga2kugel: yeah I tested a 2MB mp3
21:02:33kugeltest_codec is broken :/
21:03:00saratoga2yeah it needs to be able to rebuffer tracks
21:03:01kugelit doesn't update the display anymore when finished. if the backlight was off at this point, you cannot get the result
21:03:09saratoga2kugel: that too
21:03:16saratoga2thats the lock up thing i was talking about the other day
21:03:21saratoga2i didn't realize it wasn't frozen
21:03:36saratoga2if you turn off the backlight timeout the problem goes away
21:03:43saratoga2or just keep hitting buttons so the back light never turns off
21:04:01kugelthis also means you cannot test without constant lcd updates, which surely affects the result
21:04:32saratoga2it only updates the text about once a second, so hopefully not too much
21:05:11kugelbtw, have you seen my test_fps results?
21:05:27saratoga2from yesterday?
21:05:58 Quit motionman (" ajax IRC Client")
21:06:05kugelthey're pretty bad. only 50fps at max (boosted, full update)
21:06:13saratoga2i saw but until we have MMU running they may not mean very much
21:06:28 Quit dreamlayers ()
21:06:55saratoga2err may not mean what we think they do
21:06:59kugelI'm not sure if that will influence it
21:07:06saratoga2everything is slow right now for some reason
21:09:19 Join feisar [0] (
21:10:59 Join Zoxc [0] (
21:12:13*kugel wonders about the colums in LcdFrameRate
21:12:44kugelsaratoga2: got 140MHz with my file
21:13:25kugel128kbit/ss cbr
21:15:25 Quit ultrasonic (Client Quit)
21:16:16 Quit Zarggg ()
21:17:02kugel132MHz for one of my oggs
21:20:02saratoga2kugel: heres a neat idea: comment out the call to mad_synth_thread_ready in mpa.c
21:20:21saratoga2that will prevent mad from running the synth filterbank, which uses 50% of the CPU time
21:20:43saratoga2if decoding gets 2x faster, then the slowdown is in the codec, otherwise its some other part of rockbox
21:22:28kugelnot today anymore :(
21:23:19kugelI'm clocking CGU_AUDIO a bit higher, it might get a little faster with it (it's at the lowest clock currently)
21:24:38 Quit motionman95 (" ajax IRC Client")
21:27:12 Quit blithe ("Lost terminal")
21:27:23 Join blithe [0] (
21:36:48 Join {phoenix} [0] (
21:42:19 Quit saratoga2 ("CGI:IRC (EOF)")
21:42:46 Join miepchen^schlaf [0] (
21:44:50 Join XazZ [0] (n=XazZ@unaffiliated/xazz)
21:53:25 Quit Thundercloud (Remote closed the connection)
21:57:44 Quit feisar (Read error: 104 (Connection reset by peer))
21:59:27 Join feisar [0] (
21:59:47 Quit feisar (Remote closed the connection)
22:03:56 Join robin0800 [0] (
22:09:16 Quit robin0800 (Read error: 104 (Connection reset by peer))
22:09:39 Join bertrik_ [0] (
22:10:01 Join robin0800 [0] (
22:14:00 Join CaptainKewl [0] (
22:14:19 Quit DarkSpectrum- (Read error: 110 (Connection timed out))
22:16:17 Quit bertrik (Read error: 113 (No route to host))
22:17:04 Nick bertrik_ is now known as bertrik (
22:20:25 Join feisar [0] (
22:25:43 Quit GodEater (Read error: 113 (No route to host))
22:28:29***Saving seen data "./dancer.seen"
22:29:01 Quit MethoS- (Remote closed the connection)
22:44:36 Nick fxb is now known as fxb__ (
22:57:33 Join Seed [0] (
22:58:11 Join saratoga_work [0] (i=9803c6dd@gateway/web/ajax/
22:59:14saratoga_workkugel: you need to disconnect one of the battery leads and run it through a multimeter then into the device
22:59:19saratoga_workunfortunately this is hard to do on the fuze
23:00:01saratoga_workyou can also splice a USB cable and measure the 5v line on it
23:00:01kugelthat means opening it again
23:00:24saratoga_workthats still pretty accurate since rockbox doesn't charge the battery on the fuze
23:00:25kugelhm, I could do that
23:00:32kugelmy other cable is broken anyway
23:00:38kugelit doesn't?
23:00:51saratoga_worki don't think charging is enabled on the fuze yet
23:01:10kugelI thought it was just not calibrated
23:01:30kugelso, I might have the tools in my uni to go the cable way
23:01:35saratoga_workwiki says no
23:01:47saratoga_workMMs are really cheap if not
23:02:05saratoga_workwhat do you want to do?
23:02:29amiconnkkurbjun: Freezing where/ under what condition(s)?
23:02:40kugelget charging done, and messure savings with dbop turned off
23:03:33saratoga_workso measure how much power the OF uses charging and not charging and try to back out the charging current setting?
23:03:51 Join froggyman [0] (n=47ba40e2@gateway/web/cgi-irc/
23:04:14kugelsomething like that, yes :)
23:04:39froggymanLlorean i have a question for you
23:04:51saratoga_workthat could be a bit tricky via the USB line, since you'll have to assume an efficiency for the charger's voltage converter
23:05:04*gevaerts steals froggyman's question and runs off with it
23:05:23saratoga_worki looked online and couldn't find the exact capacity of the fuze battery, but its very close to that of the e200 judging by its size
23:05:25kkurbjunamiconn, it looks like it's related to the font/font size. If the screen height is set to 16, and it loads a non-8x8 font, the player freezes
23:05:36kkurbjunit freezes right after boot
23:05:37kugelhow would you do it? maybe I should talk to bertrik about the details
23:05:39saratoga_workso taking a value a little less then the e200 but a little more then the clip should be pretty safe
23:05:41kkurbjunand the simulator does it too
23:05:50amiconnkkurbjun: Sounds weird....
23:06:10*amiconn is about to build an mr500 sim with the remote resolution set back to 79x16
23:06:20Lloreanfroggyman: And that question is?
23:06:26kkurbjunrun it first without doing make install and it will work
23:06:35kkurbjunthen run it again after doing make install and it will freeze
23:07:02kugelsaratoga_work: bertrik said the mAH aren't very interesting for charging as is anyway, only for estimation
23:07:13froggymanhow is it that you manage to always be on this IRC and on the rockbox support forums, do you even have a job?
23:07:13saratoga_workthats true
23:07:31amiconnkkurbjun: I bet it's a list problem or something like that
23:07:35saratoga_workits mainly just so that you know how much current to give them (you shouldn't charge them with more then 1/2 their capacity per hour)
23:07:50Lloreanfroggyman: There's 146 people in this room, do you honestly believe every single one of them is at the computer?
23:07:53amiconnWith a font height > 8, th elist has no line to use when the statusbar is on
23:08:11amiconnStatusbar should probably be permanently disabled on the olympus remote
23:08:26kkurbjunthat's what I was thinking too
23:08:46kkurbjunbut I would worry that fonts larger than 16x16 would cause problems too
23:08:58kkurbjunright now the screen is running at 1/4 resolution
23:09:02Pepsi1x1i see, the reason the nano 2g isnt supported is due to crytpo, does anyone have a good resource for information on what has already been discovered in relation to that?
23:09:12froggymanwell, i guess not...
23:09:17kkurbjunbut if someone builds with a 640x480 build they will probably want to use larger fonts
23:12:53amiconnAt least the list should degrade gracefully - either just stop showing up on the remote without freezing the rest of rockbox, or probably better, just clip the font
23:13:15amiconnWe can't do more until at least basic multifont support will be added
23:13:31amiconn(basic multifont == one font per display)
23:13:49kkurbjunamiconn, don't we still build with the system font?
23:13:58saratoga_workPepsi1x1: wiki, forum thread, and that linux for nano site all had some info, i'm sure you can search and dig it up
23:14:04kkurbjunI wonder if that could be forced on the remote
23:14:07amiconnSure, but using that is worse
23:14:28kkurbjunwhy is that?
23:14:30amiconnThe sysfont only covers iso8859-1
23:27:38amiconnHmph. 'make fullinstall' is b0rked
23:28:15 Quit kugel (Read error: 110 (Connection timed out))
23:28:35kkurbjunhmm, I can't see where this is failing, it looks like the code is still running, just that the screens are no longer updating, the backlights never change. It just sits in button.c uisdl.c, and kernel-sdl.c.
23:29:43 Quit parafin (Remote closed the connection)
23:30:03 Quit bmbl ("Woah!")
23:31:50 Join parafin [0] (
23:32:00amiconnkkurbjun: I'm 99% sure this is a list problem
23:32:28amiconnI created a config.cfg manually, just setting the font
23:32:58amiconnFonts <= 8 pixel (regardless which one) and the sim works. 9 pixels -> boom
23:33:50kkurbjunyeah.., I was just trying to run through gdb and see where it's failing.. I'm not too familiar with the list code yet
23:34:21kkurbjunit just freezes for you too, no segfaults or anything like that right?
23:35:18 Quit XazZ ("Leaving")
23:35:29amiconnStatusbar disabled -> 9 pixel font works
23:37:00amiconn16 pixels work, 18 pixels -> freeze
23:38:44amiconnUnfortunately the statusbar can only be disabled for both displays at once. While I do think that it should be disabled permanently on the mrobe remote, it would probably make sense to have a separate setting for main & remote on other targets with remote lcd
23:39:32kkurbjunyeah, agreed, that's what I was thinking of doing eventually
23:41:52 Quit {phoenix} (Remote closed the connection)
23:42:33 Join hillshum [0] (n=hillshum@unaffiliated/hillshum)
23:43:31 Quit faemir (Read error: 104 (Connection reset by peer))
23:43:36 Join Bernd [0] (n=5b277c86@gateway/web/cgi-irc/
23:46:44 Join taylor_ [0] (
23:47:30 Part wincent ("Kopete 0.12.7 :")
23:50:09 Join {phoenix} [0] (
23:51:07 Quit Lss (Read error: 54 (Connection reset by peer))
23:51:12 Join Riku [0] (
23:52:14 Quit robin0800 (Read error: 104 (Connection reset by peer))
23:52:16soapLlorean: i just realized one of the other ideas we had discussed re:the forums, was to move "Hardware" and "Repairing Broken Players" to what has now become "Third Party"
23:52:37saratoga_workdisabling the synth filter bank on the Fuze for Mp3 makes MP3 much faster, but still takes ~80MHz
23:52:51soapand while doing so to unnest "Repairing Broken Players" from its "hiding place" under hardware.
23:54:08soapNot because the Hardware forums have ever been "out of control" but because we strengthen the line between what is Rockbox (and we can support to the best of our abilities) and what is Not Rockbox and we mearly tolerate. ;)
23:54:21Lloreansoap: Well "Hardware" shouldn't be third party
23:54:34soapIt isn't Rockbox. ;)
23:54:40LloreanIts intent, at least when I originally organized things, was to be basically the "firmware" layer of Rockbox
23:54:57LloreanQuestions about battery calibration, or other issues where it's Rockbox's support of the hardware.
23:55:13soapas in questions regarding support for hardware not discussion of replacing batteries and such?
23:55:17saratoga_workso it looks like the CPU is very slow (which is odd since I've got nearly everything in IRAM), but other things are probably also very slow (which makes more sense, maybe buffering is also very slow without cache)
23:55:17LloreanThe repairing broken players on the other hand is a gray area, since arguably nothing there actually directly involves Rockbox
23:55:26Lloreansoap: Exactly. Replacing batteries is "repairing"
23:55:33saratoga_workat least we know theres not one device driver busy waiting 90% of the cpu's time
23:56:31Lloreansoap: I'll reorg Hardware and Repairing into the originally intended Hardware (under Rockbox) and "Repairing / Upgrading Players"
23:56:42LloreanWith some new POSTING IN THIS FORUM and descriptions.
23:56:56LloreanThen hardware can stay under Rockbox, and Repairing/Upgrading can go to third party
23:57:35soapI'm thinking about that "POSTING IN THIS FORUM" - because I see what you're getting at in this context, I didn't see before, and I'm trying to figure out why I didn't grok it before.

Previous day | Next day