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 2014-01-08

00:01:48 Quit lorenzo92 (Ping timeout: 240 seconds)
00:04:53foolshDid I get banned from the forum by accident or am I being punished for something I don't remember doing? (username BenBrown)
00:06:01*gevaerts checks
00:07:11gevaertsLooks like saratoga got too trigger-happy at some point
00:08:06gevaertsUnfortunately unbanning is as slow as banning...
00:09:15*gevaerts suspects saratoga just went over deleted posts and didn't pay enough attention at the crucial moment
00:09:37kugelamiconn: ping
00:09:48kugelamiconn: the 2bit lcd drivers are buggy
00:10:00kugelthey don't account for the drawmode in lcd_bitmap_part()
00:11:45foolshgevaerts: Okay but a life time ban seems kinda harsh for clicking "post" instead of "save" during an edit :-)
00:12:15gevaertsfoolsh: the thing is that the other deleted posts around it were actual spam :)
00:12:27gevaertsAnd a lifetime ban for that is *mild*
00:12:44gevaertsAnyway, with any luck you should be back in
00:13:56gevaertsHmm, you'll probably have to wait a few more minutes. I seem to be keeping the forums busy again
00:14:19kugelamiconn: FS #12927
00:14:20fs-bluebot 32bit lcd drivers ignore drawmode in lcd_bitmap_part() (bugs, new)
00:14:52kugelthe fancy bit shuffling magic in there is a bit above my head right now to fix it myself
00:17:33 Quit lebellium (Quit: ChatZilla [Firefox 27.0/20131216183647])
00:21:43gevaertsfoolsh: all done now!
00:21:55foolshgevaerts: Thank you,
00:22:59gevaertsNo problem. Sorry about the ban
00:24:09foolshgevaerts: Do you think you could help me out? Me being the novice can't see why I get (undefined reference to `strlen' ) in g#710
00:24:12fs-bluebot3Gerrit review #710 at : 3Preliminary proof of concept SHA256 function by Benjamin Brown (changes/10/710/1)
00:26:02gevaertsI'll have a look, but I'm no plugin specialist
00:26:59saratogafoolsh: sorry, I deleted your duplicate post at the same time as some spam and apparently clicked the wrong one when banning
00:27:19saratoga(they both show up together in the deleted posts subforum)
00:27:49foolshsaratoga: That's alright
00:28:39gevaertsfoolsh: you need rb->strlen
00:29:07saratogaabout that bitcoin thing, it wasn't clear to me, does the machine have to be actively networked to recieve coins, or can you somehow copy the hashes back and forth without that?
00:31:08foolshNot sure, but I think its more like a one time use code or hash that will payout from your unnetworked wallet to a networked wallet.
00:32:15[Saint]either or, frankly, its a terrible idea. :)
00:33:13foolsh[Saint]: Absolutely it is, but if pays, its great idea to me ;-)
00:33:13saratogaso if a wallet is unnetworked, what happens to coins that are sent to it? can they be redeemed at any point in the future?
00:34:55copperthe money is in the blockchain I think
00:35:15coppera wallet is just a set of private keys unlocking the money in the blockchain
00:35:16[Saint]I dunno, I see your point, but I like to think I have a moral obligation to not implement clearly terrible ideas, especially when there are readily available tools for the same purpose on sane platforms.
00:35:41[Saint]But I see the point, money is money.
00:35:41saratogai think its an impractical idea, but i don't see any harm in it if someone wants to use it
00:35:43coppersaratoga: yes you "get" the money even if your wallet isn't networked
00:36:09copperthe "money" is really just a big fat log of transactions
00:36:34saratogaso in practice this thing would only ever be used when sending or claiming coins?
00:37:02 Quit pamaury (Ping timeout: 240 seconds)
00:38:38copperto safekeep the private keys, and later to conduct transactions, yes
00:39:20[Saint]well..."safekeep", in this instance.
00:39:29[Saint]We should probably just say "store". :)
00:39:36coppererm, yes
00:40:29copperand to generate addresses (pairs of public and private keys)
00:40:30[Saint]physical access to the device by a determined individual would result in a very bad day.
00:41:29copperoh and, the log of transactions (the blockchain) is public, and shared on a peer to peer network
00:41:34copperthat's the actual money
00:41:45 Quit kugel (Ping timeout: 260 seconds)
00:42:01copperit just says "money went from this address to that address"
00:42:21copperand everyone recognizes that address X has a balance of Y BTC
00:42:42copperand the guy with the private key to that address is able to tell the network that he wants to "send" money to address Z
00:43:00copperwhich the network will also confirm
00:43:24copperwherever the private key is, no-one can change the balance of an address without it
00:43:42copperand the information about the balance of that address is perpetually shared globally
00:44:49copperif the private key gets lost, the "money" is lost forever
00:45:05foolsh[Saint]: I think the sane way to use such a plugin would be to create a wallet not associated with your main account and transfer a small amount of funds, as needed. That's the way some do it now any way, just to stay safe. Like not putting all your cash in your wallet when you go out.
00:45:15copperso if that Rockbox plugin crashes during a USB transfer and fucks up the private keys…
00:45:51copperUSB instability with Rockbox would be my primary concern here
00:46:20copperand, having users lose money would open a whole new can of worms for Rockbox
00:46:37[Saint]foolsh: Obviously its a differing of opinion, but given the platform - and the fact that there are so many implementations for the same thing on sane, everyday platforms, I don't think there is a sane way to use this.
00:47:02[Saint]There's a way to use it. But, I wouldn;t necessarily call it sane. :)
00:47:19foolsh[Saint]: Absolutely correct in every way :-)
00:47:48foolshgevaerts: Thanks again that got it linking properly
00:48:20*[Saint] thinks the obvious name for this plugin is "CoinBox", but cringes.
00:50:57***Saving seen data "./dancer.seen"
00:52:31[Saint]Hmmm...I wonder if you could use a bunch of random audio files as entropy.
00:53:43gevaerts[Saint]: how will you select the random audio files?
00:54:26[Saint]duh...randomly. ;P
00:54:41[Saint]No, seriously, good question. I hadn;t thought that far.
00:54:57gevaertsRight, so you're not actually solving anything :)
00:55:28JdGordonusing specific audio files to seed a prng is a good way to make a password acually
00:55:40JdGordonas long as you don't tell anyone which file you used
00:56:47saratogaor just delete it once its seeded
00:58:16 Quit ender` (Quit: I believe in free will, but I don't really have a choice.)
00:59:32*[Saint] supposes the pseudo-random playlist generation used for the playback engine would be good enough to pick random files for entropy
01:00:29[Saint]unless you only had a single audio file or so.
01:18:19 Part foolsh
01:19:09 Quit Zarggg (Quit: Zarggg)
01:29:01saratogalot of improvements to the official opus decoder lately
01:39:30[Saint]Any we really care about?
01:40:12saratogayeah, should be faster and use less memory/stack
01:42:26[Saint]How difficult (or not) is it to rebase their upstream stuff on top of our codebase?
01:48:27 Join ikeboy [0] (
02:07:52 Quit chrisjj (Ping timeout: 272 seconds)
02:50:59***Saving seen data "./dancer.seen"
02:56:39 Quit ikeboy (Quit: ikeboy)
03:15:23 Quit dfkt (Remote host closed the connection)
03:31:19 Quit scorche (Ping timeout: 245 seconds)
04:01:34 Quit amiconn (Disconnected by services)
04:01:34 Quit pixelma (Disconnected by services)
04:01:35 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:01:35 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:01:37 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:01:39 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:47:21 Join scorche [0] (~scorche@rockbox/administrator/scorche)
04:51:00***Saving seen data "./dancer.seen"
04:59:06 Quit amiconn (Disconnected by services)
04:59:06 Quit pixelma (Disconnected by services)
04:59:07 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:59:07 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:59:09 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:59:11 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
05:06:46 Quit tertu (Ping timeout: 272 seconds)
05:07:03 Quit TheSeven (Disconnected by services)
05:07:17 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:11:05 Join tertu [0] (~tertu@
05:14:11 Join Strife89 [0] (
05:21:50 Quit [Saint] (Read error: Connection reset by peer)
05:22:54 Join [Saint] [0] (~saint@rockbox/staff/saint)
05:23:18 Quit tertu (Ping timeout: 246 seconds)
06:18:59 Join cmhobbs [0] (
06:18:59 Quit cmhobbs (Changing host)
06:18:59 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
06:51:01***Saving seen data "./dancer.seen"
06:53:06 Join tertu [0] (~tertu@
07:05:39 Join Rower [0] (
07:06:58 Join ter2 [0] (~tertu@
07:08:41 Quit tertu (Read error: Connection reset by peer)
07:53:08 Quit Guest39205 (Quit: Bye bye)
07:57:23 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
08:01:09 Join LinusN [0] (
08:04:19 Join wodz [0] (
08:06:21wodzlots of daps have adc with floating input - thats pretty good source for seed of some pseudo random function
08:09:18 Join ender` [0] (
08:47:00 Quit fragilematter (Quit: Leaving.)
08:51:02***Saving seen data "./dancer.seen"
08:56:20fs-bluebotBuild Server message: 3New build round started. Revision 51d8a45, 243 builds, 33 clients.
08:59:45fs-bluebotBuild Server message: 3Build round completed after 206 seconds.
09:01:34 Quit Rower (Read error: Connection reset by peer)
09:01:49 Join Rower [0] (
09:08:06 Quit Rower (Ping timeout: 260 seconds)
09:10:49 Quit wodz (Quit: Leaving)
09:15:00 Quit scorche (Disconnected by services)
09:15:03 Join scorche` [0] (~scorche@rockbox/administrator/scorche)
09:18:27 Quit ter2 (Ping timeout: 265 seconds)
09:21:22 Join lebellium [0] (~chatzilla@
09:22:49 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
09:33:47 Join kugel [0] (
09:33:47 Quit kugel (Changing host)
09:33:47 Join kugel [0] (~kugel@rockbox/developer/kugel)
09:44:55 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:07:41 Join wodz [0] (
10:16:10 Quit mt (Ping timeout: 260 seconds)
10:20:08 Join Zagor [0] (~bjst@
10:20:08 Quit Zagor (Changing host)
10:20:08 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
10:29:05 Quit pamaury (Read error: Operation timed out)
10:36:29pixelmathe delta table looks a bit broken here (too short)
10:37:50*gevaerts bets Zagor is the right man to have a look :)
10:38:23 Quit Raptors (Quit: Leaving)
10:39:04wodzgevaerts: If the logf() buffer is cropped as you said I think the change is good idea
10:40:02gevaertswodz: there's a slight issue with it not correctly merging the two strings at each end of the ring buffer on wrap though
10:40:38gevaertsBut yes, I think it's useful and doesn't hurt
10:41:27wodzI think we should also resurrect pamaury's patch to print function names in backtrace (for DEBUG builds)
10:41:36kugeldoes logf use the log-to-disk framework already?
10:44:18wodzI mean g#208 after some tweaks
10:44:19fs-bluebot3Gerrit review #208 at : 3 by Amaury Pouly (changes/08/208/1)
10:44:57gevaertskugel: not as far as I know. I'm also not convinced that it should, though
10:46:15fs-bluebotBuild Server message: 3New build round started. Revision 124e9c1, 243 builds, 32 clients.
10:46:51wodzgevaerts: why?
10:47:28kugelyea, why?
10:47:37gevaertsBecause as I see it logf is for the sort of debugging where you have to be careful with triggering side effects
10:48:03gevaertsCurrently logf() is interrupt-safe. Will that still be the case if you make it use log-to-disk?
10:48:36kugelhm, good point
10:48:43 Quit fs-bluebot (Remote host closed the connection)
10:48:58 Join fs-bluebot [0] (
10:49:11 Quit fs-bluebot (Remote host closed the connection)
10:49:11wodzI think log-to-disk logs to intermediate ram buffer and dumps to disk when the disk is spinned up.
10:49:25 Join fs-bluebot [0] (
10:49:25fs-bluebotBuild Server message: 3Build round completed after 182 seconds.
10:49:44gevaertsAlso, sometimes you're debugging disk drivers :)
10:50:10wodzyes, thats the only point I can think of to not integrate the two
10:50:24kugelthat's no excuse for degrate logf as a whole
10:50:28gevaertsI don't mind integration as such, but we do have to be careful
10:51:03***Saving seen data "./dancer.seen"
10:51:32wodzmaybe on logf() builds we should disable callback dumping buffer to the disk and let it be manual operation as is in logf() now?
10:52:24 Join petur [0] (~petur@rockbox/developer/petur)
10:52:32kugelyea, logfdisk doesnt trigger disk access on its own, only when the disk spins up for another reason
10:53:10kugelso logf remains interrupt safe
10:54:39kugelexcept when register_storage_idle_func() is configured to call the callback syncronously, i.e. for BOOTLOADER and APPLICATION
10:54:41gevaertsWe could also just have an advanced configure option that disables the fancy stuff
10:54:56 Nick SuperBrainAK is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
10:55:22gevaertsUsually you know what sort of thing you're going to be debugging before you do this sort of build anyway :)
11:06:49 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:10:08pamaurywodz: I don't remember the cost of the backtrace patch to print names, it requires the compiler to add a string for every function, it's potentially expensive
11:10:26pamauryit should probably only be enabled in debug mode or on targets with lots of memory
11:11:58wodz :-)
11:13:36pamauryhowever for users to report bugs, having names a huge win
11:20:16pamauryjust checked, on a typical imx233 build it's cost about 80 KiB for the main firmware
11:21:30pamaurythere are others possibilities though, but more complicated ones, like clever compressing of the strings and then put them at some place in the binary
11:22:04pamauryI made some experiments at the time, but I don't remember the results, if someone has the logs maybe he/she could grep for it
11:38:37 Quit pamaury (Ping timeout: 240 seconds)
11:43:06 Join foolsh [0] (
11:49:50 Quit fragilematter (Quit: Leaving.)
11:50:35 Quit lebellium (Read error: Connection reset by peer)
11:52:25 Join lebellium [0] (~chatzilla@
11:59:15 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
12:01:18 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
12:01:29PurlingNayukiJdGordon: G697 updated
12:01:31fs-bluebot3Gerrit review #697 at : 3Implement volume limiter feature by PurlingNayuki (changes/97/697/15)
12:03:56PurlingNayukiAnd this time is without a WIP prefix
12:14:19 Quit Zagor (Read error: Operation timed out)
12:16:18 Quit pamaury (Ping timeout: 246 seconds)
12:31:59 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
12:32:04 Join Scall [0] (~chat@unaffiliated/scall)
12:35:23 Join ikeboy [0] (
12:44:53JdGordonPurlingNayuki: and commented
12:44:56JdGordonand gone! o/
12:50:19 Quit ikeboy (Read error: Connection reset by peer)
12:51:05***Saving seen data "./dancer.seen"
12:51:21 Join ikeboy [0] (
12:52:40PurlingNayukiJdGordon: I'm sorry the default should be obviously and is the max volume since the first patchset
12:53:03PurlingNayukiI made it wrong in the changelog
12:56:24PurlingNayukiAny example for using CUSTOM_SETTING()? I didn't notice it before.
12:57:18 Join Raptors [0] (
13:02:59 Quit Raptors (Read error: Operation timed out)
13:06:45 Join Raptors [0] (
13:11:46PurlingNayukiOK I found one in EQ settings.
13:16:26 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
13:24:21 Quit ikeboy (Quit: ikeboy)
13:25:44 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:29:45 Nick funman_ is now known as funman (
13:53:16 Join chrisjj [0] (561bb732@gateway/web/freenode/ip.
13:53:29gevaertsZagor: did you see in the backlog that the rbutil source got mixed up with one of the binaries?
13:53:40gevaerts doesn't have the 1.4 source
14:00:54Zagorgevaerts: fixed
14:06:04chrisjjAnyone know which patch lost the WPS Playlist Viewer Settings command?,46036.0.html
14:06:12 Join mt [0] (
14:10:14 Join pretty_function [0] (~sigBART@
14:17:00chrisjjkugel: <perhaps?
14:22:41 Quit fragilematter (Quit: Leaving.)
14:22:49 Quit onder` (Ping timeout: 272 seconds)
14:26:37 Quit lebellium (Quit: ChatZilla [Firefox 27.0/20131216183647])
14:28:20alucrydZagor, thx, now building :)
14:36:26gevaertsJdGordon: you didn't update the manual after 97a4c1ef
14:51:06***Saving seen data "./dancer.seen"
15:02:51 Join amayer [0] (
15:08:55 Quit cmhobbs (Read error: Operation timed out)
15:19:49 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
15:25:55 Quit wodz (Quit: Leaving)
15:29:20 Part LinusN
15:40:32 Quit Cultist (*.net *.split)
15:43:07 Join Cultist [0] (~CultOfThe@
15:46:54 Quit pretty_function (Remote host closed the connection)
15:47:56 Join MarcAndersen [0] (
15:48:40MarcAndersenhi, I just wanted to know if the zen x-fi can charge now without going into recovery?
15:52:08chrisjjYes it can.
15:53:32chrisjj(Version -131217)
15:53:59MarcAndersenwow, thanks
15:54:20MarcAndersenbut it's not stated at the port wiki page though
15:54:21chrisjjHowever I think it does now crash more often than before.
15:54:36MarcAndersenyeah i think so too
15:55:47 Join meehoo [0] (5e4865e2@gateway/web/freenode/ip.
15:55:54meehoohi there
15:56:12 Quit mt (Ping timeout: 272 seconds)
15:56:14chrisjjThe author has said he hasn't implemented charging. I guess that's the reason he didn't mention charging on the port page.
15:56:38meehooI've a problem with the latest rockbox, while recognizing my sansa clip plus it crashes.
15:57:08meehooevery tie
15:57:17gevaertsmeehoo: you mean rockbox utility 1.4.0?
15:58:03gevaertsWhich operating system are you using?
15:58:59gevaertsOK. For now I'd recommend using an older version, which you can get at
15:59:16gevaertsbluebrother: I'll leave investigating this to you
16:01:13 Quit meehoo (Quit: Page closed)
16:01:15chrisjjMarcAndersen: I've added a note about charging to the port page. If you verify changing works on your unit too, please note that as well.
16:01:38 Join meehoo [0] (5e4865e2@gateway/web/freenode/ip.
16:01:55MarcAndersenIs there any plans about the next release of rockbox?
16:02:27chrisjj"I guess that's the reason he didn't mention charging on the port page." Correction: ... mentioned not charging on the port page :)
16:03:32meehoo Now I dont know how to change my player into the MSC mode
16:03:36toehser1Something happens sometimes where my theme will suddenly "lose it's font" and the font will switch to rockbox default font and then a minute later it crashes. This is on a Clip Zip with a pretty recent build. It doesn't only happen when a USB cable is involved. It _does_ seem to be related to only when certain MP3 files are playing - one playlist it never happens, another it happens pretty soon. I get things like "Undefined instru" but also other cr
16:04:04meehooi's requiered for the furher installation of rockbox
16:05:22chrisjjtoehser1: does the crashing track happen to show a very long scrolling text item on the WPS?
16:06:30gevaertsmeehoo: says what to do
16:06:35toehser1Probably, yes, I have some long scrolling concatenated fields.
16:06:49meehoothanks evilnick
16:06:56meehoothanks gevaerts
16:07:51toehser1Is there a known text-scrolling bug you can point me to?
16:08:49chrisjjtoehser1: that's known to cause crash under certain circumstances. A smaller font might alleviate.
16:10:25MarcAndersenwell, what do you mean with Rockbox 3.14 was released on XX XXXXX 2014? it would be better not to write "was" in there :)
16:11:10gevaertsMarcAndersen: it means it's a draft document
16:11:34gevaertsIt's not released yet, but when it is, that date will be changed
16:11:41MarcAndersenbut you have not frozen the build yet?
16:12:08toehser1Urrr... known as in "huh, we saw that, no-one has filed a bug or investigated" or known as in "there's a bug # but no-one has dug into it" or known as in "maybe it is in such and such a function in xxx.c" or known as in "we think we know what causes it but haven't fixed it"... Does it fail in the simulator also? Is there a doc on how to go from the crash screen back to where I can be in a source debugger? Are there debug options I should set to try to
16:12:10chrisjjtoehser1: Sorry, no. I've not seen this bug filed.
16:12:15 Quit Strife89 (Ping timeout: 252 seconds)
16:12:34chrisjjMarcAndersen: Agreed. I didn't write it.
16:12:47 Join pretty_function [0] (~sigBART@
16:12:52toehser1Well, I can file a bug and start to investigate. Something to fix more interesting that what they give me at work would do me good.
16:14:09gevaertsMarcAndersen: we're not frozen yet, no
16:14:12toehser1It happens with a fairly small font, but it does seem that there are always a few lines long enough to scroll, so that is for sure seeming like a hopeful lead.
16:15:17toehser1All the fonts are anti-aliased ones, if that might make a difference.
16:15:33chrisjjtoehser1: as in #1 :)
16:15:49 Join ter2 [0] (~tertu@
16:15:49 Quit einhirn (Ping timeout: 264 seconds)
16:16:27toehser1I was hoping for "known as in #4" :)
16:16:47chrisjjtoehser1: I can imagine it would make a difference :) Good info!
16:17:11chrisjj("it" == "antialiased")
16:18:32 Quit foolsh (Quit: foolsh)
16:18:39toehser1I'll try copying the playlist that fails to where I have the sim built from git, and see if it fails there, too. I don't suppose there is some trivially simple way to attach a remote source level debugger to it and set breakpoints?
16:18:51meehooits not clear for me as to how to change the player into different USB modes,where's the place to do it ??
16:19:35toehser1I don't think Rockbox has more than one USB mode, but on my player the native firmware supports both MTP and MSC, if that is what you mean.
16:20:01meehooI am supposed to change my player into MSC mode. don't know how to do it
16:20:11toehser1what player
16:20:16meehoootherwise I am unable to install the Rockbox
16:20:44meehooI have sansa clip +
16:21:42toehser1I think it is under something like "system menu"
16:21:54toehser1not too hard to find, those menus aren't that deep
16:22:05meehoobut, where are they?
16:25:11toehser1Well, I don't have a clip+ in front of me, and it has been a while, but, what I was trying to say is, roughly, I don't know where they are, but they aren't hard to find, if you walk the menu tree. Essentially, that means, descend each branch of the menu tree, looking for it... I could google it for you?
16:25:39toehser1This isn't a rockbox question, after all, so any Sansa Clip+ google search for, say, "sansa clip+ how to enable MSC" is likely to find it.
16:26:40toehser1Google says you can force it by connecting it to the PC with the center button pressed.
16:26:41meehoook, but where are those menu trees, in the rockbox utility or in windows
16:27:03toehser1In the SANSA. Boot the Clip+, and navigate with the buttons on the player.
16:27:13toehser1NOT in the rockbox utility and NOT in Windows. On the player.
16:27:14meehooI can browse my sansa clip built-in disk with the explorer
16:27:23meehoobut its a different mode than MSC
16:27:27toehser1as a drive letter or as the MTP?
16:27:33toehser1you have to disconnect it from the computer
16:27:38toehser1then use the menu on the player
16:27:47toehser1there is an MTP/MSC option in the menus on the player
16:27:55meehooah, great
16:28:05toehser1Or, there is the force-with-center-button way, from power-off
16:28:32toehser1The center button way, it has to be powered off, then you connect it while holding the center button
16:28:38toehser1but I would use the menu instead.
16:29:59meehooI've just checked the settings, it was switched to MSC mode already
16:30:13meehoostill windows recognized it as somehing different
16:31:06toehser1What behavior makes you think Windows isn't seeing it as MSC?
16:31:07meehooI'll try with the r. utility now
16:31:40meehoothe rockbox utility says it has to changed into MSC
16:33:19meehoonow it works
16:33:39meehoobut now I have to load a bootloader file
16:34:00meehoois it to be downloaded from the site?
16:35:19meehoofor I dont have a bootloader file
16:37:21toehser1There's a link around for the bootloader. I think the download comes directly from Sansa for IP reasons.
16:38:53 Quit fragilematter (Quit: Leaving.)
16:39:07 Quit kugel (Read error: Operation timed out)
16:40:18 Join foolsh [0] (
16:40:38 Part foolsh
16:41:53chrisjjAnyone know what keypress exits the simulator on Windows?
16:51:10***Saving seen data "./dancer.seen"
16:57:13 Quit pretty_function (Remote host closed the connection)
16:58:27 Join lebellium [0] (~chatzilla@
17:01:42 Quit Zagor (Quit: Clint excited)
17:12:46meehootoehser1 : thsnk for help. my rockbox is on board
17:12:56 Quit meehoo (Quit: Page closed)
17:19:22 Join foolsh [0] (
17:24:07 Join meehoo [0] (5e4865e2@gateway/web/freenode/ip.
17:24:29meehooshit, my sansa clip froze
17:25:23meehooeven if I unplug it from the pc its still frozen
17:25:35meehooany idea on how to unfreeze it?
17:25:51gevaertsHold power for at least 30 seconds for a hard reset
17:26:50meehoothanks a lot
17:30:15 Join lebellium_ [0] (~chatzilla@
17:31:38 Quit lebellium (Ping timeout: 260 seconds)
17:31:39 Nick lebellium_ is now known as lebellium (~chatzilla@
17:36:19 Quit petur (Quit: bye!)
17:37:56 Join kugel [0] (~kugel@rockbox/developer/kugel)
17:39:03 Join pretty_function [0] (~sigBART@
17:45:33 Join Rower [0] (
17:47:36pamaurychrisjj: either use the "power" button just like with a normal device, or hit Alt+F4 or close the window
17:52:53 Quit ter2 (Read error: Connection reset by peer)
17:53:06 Join ter2 [0] (~tertu@
17:58:50 Join n1s [0] (
17:58:50 Quit n1s (Changing host)
17:58:50 Join n1s [0] (~n1s@rockbox/developer/n1s)
18:01:25pamaurychrisjj: I have corrected ZEN and X-Fi wiki w.r.t to charging: it is supposed to work in the current (not officially released though) bootloader
18:03:55foolshpamaury: What do I need to do to g#677 to get it merged, only create a touchpad settings submenu? or try to add a deadzoned button matrix for touchscreen.c and button-fuzeplus.c somehow?
18:03:56fs-bluebot3Gerrit review #677 at : 3Fuze+: add a configurable deadzone area for touchpad buttons by Amaury Pouly (changes/77/677/12)
18:06:04chrisjjpamaury: There's no power button showing on the Clip Zip simulator. A click at the position of the power button does nothing.
18:06:39pamauryall targets have keys for all buttons, though they might not be documented and need to be looked up in the source code
18:06:48pamaury*all targets should have
18:07:02chrisjjpamaury: ALT+F4 doesn't work, and hasn't worked for me for four years on every simulator I've tried:
18:07:23pamauryyou can close the window, this should always work and works for me
18:07:32chrisjjDoes simulator ALT+F4 work for you? Or for anyone else here?
18:07:56 Quit mc2739 (Ping timeout: 272 seconds)
18:08:17pamauryno idea for ALT-F4, it was a guess
18:08:40chrisjjNot a bad guess :)
18:09:31chrisjjpamaury: "or close the window". That's what I'm trying to do - by keyboard.
18:09:35 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
18:10:57gevaertsSimulator keymaps are all in uisimulator/buttonmap/*
18:11:11pamauryfoolsh: it's unclear, a touchpad/screen submenu would definitely be a good thing. Add support for touchscreen deadzone is not that easy. It also raised the interesting following question: how do you handle different screen sizes ? I think the best answer (which could apply to the fuze+) would be to measure the deadzone size in percentage of the button size so that it applies nicely.
18:11:28pamaurykugel: any opinion on this ? I seem to recall you have done some work on touchscreen
18:13:30 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
18:14:39 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:19:25chrisjjgevaerts: Thanks. Seeing "case SDLK_ESCAPE: ... new_btn = BUTTON_POWER" hinted than Log Power might exit, and it does, so problem solved.
18:21:04chrisjjpamaury: Re ZEN, yes I see your "Charging works". Thanks.
18:25:12bluebrothermeehoo: which version of Rockbox Utility are you using? Do you already have 1.4.0?
18:25:34bluebrotherit shouldn't crash with 1.4.0 anymore, but I don't have the device to try that myself
18:26:26 Nick DormantBrain is now known as SuperBrainAK (~andy@2001:470:8:a61::5f92:59a1)
18:34:57foolshpamaury: Something like this? Add deadzone support in touchscreen.c (based on a percentile) and rewrite button-fuzeplus.c to use the SAME exported deadzone value from a touchpad/touchscreen deadzone setting, That has it's own submenu?
18:40:47pamauryyeah, both touchpad and touchscreen would need to use the same function to compute the button/deadzones. If you are not in a hurry I can have a try this week end
18:42:38foolshNo I'm in no hurry, I wasn't suggesting you do it all. Just want something to do, but anything I do will most likely have to be rewritten ;-)
18:44:54pamauryI know, that's why I'm suggesting it :)
18:47:46foolsh:-D It's still a great learning experience
18:51:12***Saving seen data "./dancer.seen"
18:52:41pamauryfoolsh: ok, if you want to try. First we need to create a new file which will serve as a library for both touchscreen and touchpad (touchscreen.c is only compiled for touchscreen), let's say touchlib.c
18:52:41pamauryFor the moment, let's say it would contain a single function like "int touch_to_grid(int **grid, int col, int row, int x, int y, int w, int h, int deadzone)" which would take an array (grid) representing a col by row grid giving the button mapping. The touch coordinate is (x,y) and the total zone size is (w,h), the deadzone is a percentage of the button size
18:53:29pamauryFor the moment you could ignore the percentage and mostly reuse the one from the fuze+ button drive in gerrit (which some tweaks). Then you need to fix touchscreen.c to use and the fuze+ driver too
18:53:54pamauryThen modify the setting to reflect that is is a percentage and make a submenu
18:55:54 Part meehoo
18:59:39foolshpamaury: "There is no 'try', only do or do not" - Master Yoda
19:00:01foolshI'll give it a shot thanks
19:01:13 Quit chrisjj (Ping timeout: 272 seconds)
19:08:05 Join terrapin [0] (6c334ac2@gateway/web/freenode/ip.
19:08:34terrapinhi there, sort of a newb question:
19:09:34 Join anewuser_ [0] (
19:09:38 Quit kugel (Ping timeout: 260 seconds)
19:11:25terrapini'm trying to install rockbox on my sansa clip+ 8gb. I've downloaded the utility and am being prompted to inform it as to which type of device i'm using. i'm in the "configuration" window, attempting to give the utility the file path for the device. however, whenever i click on the drop-down menu under the "select your device in the filesystem" label, no menu opens. it doesn't open any sort of filesystem for me to select the device
19:11:52terrapinthis is windows 7, by the way
19:13:11anewuser_reading release notes for 3.14 did it come out?
19:20:30 Quit Raptors (Read error: Connection reset by peer)
19:20:46 Quit pamaury (Ping timeout: 252 seconds)
19:20:48 Join Raptors [0] (
19:25:36 Quit terrapin (Quit: Page closed)
19:32:49 Join terrapin [0] (6c334ac2@gateway/web/freenode/ip.
19:35:33 Quit krnlyng (Ping timeout: 246 seconds)
19:37:35terrapinsorry to repost this; i decided to try restarting to see if it fixed my problem, which it unfortunately didn't. i'm trying to install rockbox on my sansa clip+ 8gb. i've downloaded the utility and am attempting to inform it of the type of device i'm using. whenever i try the autodetect feature, the program crashes. i'm now at the page titled "configuration".
19:38:39terrapin i'm trying to provide the file path for the device to the utility, which it seems should be done through the dropdown menu under the "select your device in the filesystem" label. however, whenever i click the bar to open the drop-down for the filesystem, nothing happens. i can't provide the file path for the device. any ideas?
19:42:55terrapinjust to make things a little more clear: this is the point i'm at: When i click the bar or arrow to the left of the "refresh" button, nothing happens, yet the program is definitely still running and not crashed/frozen
19:53:17 Join krnlyng [0] (~liar@
19:57:35 Quit Rower (Ping timeout: 245 seconds)
19:58:24 Join Rower [0] (
20:02:41 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
20:12:21 Quit pretty_function (Remote host closed the connection)
20:18:13 Join petur [0] (~petur@rockbox/developer/petur)
20:36:48 Quit y4n (Quit: PANTS OFF!)
20:37:54 Join meehoo [0] (5e4865e2@gateway/web/freenode/ip.
20:39:54 Part meehoo
20:51:16***Saving seen data "./dancer.seen"
20:59:57 Join kugel [0] (~kugel@rockbox/developer/kugel)
21:04:09 Quit krnlyng (Ping timeout: 252 seconds)
21:19:04 Join krnlyng [0] (
21:41:33toehser1Guessing that maybe with the scrolling happening in another thread, I should start by looking for stuff in fonts that isn't threadsafe?
21:42:23gevaertstoehser1: if you're looking at that, remember that threading in rockbox isn't preemptive
21:48:06toehser1Is there a doc that can make "pc:3004ADD0" useful as far as locating the source line? Something in the build output?
21:48:51gevaertsOr by hand, using objdump −−disassemble rockbox.elf
21:49:53 Join Tomwi [0] (
21:55:05 Quit zoktar (Ping timeout: 245 seconds)
21:58:27toehser1I have yet to make it fail in the simulator. Does that narrow anything?
21:58:42MarcAndersenHi, i have a daisy player called a victor stratus, which is running a kind of unix kernel. I can unpack the squashfs file of it's firmware and it may be possible to recreate an image and make it to update from it, what do you think?
22:01:54toehser1Though I have seen "segfault core dumped" in the simulator... where does it put that core? It says "core dumped", but I don't see a core dump...
22:02:43gevaertsWhat does "ulimit -c" say?
22:03:14toehser1ah. thx
22:23:06 Join zoktar [0] (~zoktar@unaffiliated/zoktar)
22:28:07 Join mt [0] (
22:29:39 Quit fs-bluebot (Ping timeout: 265 seconds)
22:31:05 Join fs-bluebot [0] (
22:44:19 Quit ter2 (Read error: Connection reset by peer)
22:44:38 Join ter2 [0] (~tertu@
22:50:18 Quit ter2 (Ping timeout: 260 seconds)
22:51:18***Saving seen data "./dancer.seen"
22:59:29 Quit n1s (Quit: Ex-Chat)
23:00:14 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:10:48 Quit benedikt93 (Quit: Bye ;))
23:12:27JdGordonPurlingNayuki: hey
23:12:40 Quit Rower (Read error: Connection reset by peer)
23:12:55 Join Rower [0] (
23:15:52 Join chrisjj [0] (561bb732@gateway/web/freenode/ip.
23:18:15kugelJdGordon: hey. took your 9patch function to draw a nicer splash background
23:18:28[Saint]chrisjj: that would be JdGordon, I believe.
23:18:33kugelincluding transparent drop shadow
23:18:46[Saint]The manual seems to have been broken ever since the playlist viewer was committed.
23:18:58[Saint]...which sucks.
23:20:31chrisjjJGordon: recap : Anyone know which patch lost the WPS Playlist Viewer Settings command?,46036.0.html
23:21:00[Saint]chrisjj: also, fwiw, "a very long scrolling line in the wps" is not known to cause anything.
23:21:10[Saint]Not sure where you got that from. But, its not true at all.
23:21:35[Saint]Nor will changing the font alleviate it.
23:21:53chrisjj"Not sure where you got that from". Testing.
23:22:06[Saint]Well, that is purely anecdotal.
23:22:15[Saint]Your experiences are arbitrary.
23:22:21[Saint]That doesn;t make it "known" to the project.
23:23:05chrisjjI didn't say it was known to the project. I said it was known. And I recall at least one other arbitrary user who has reported it.
23:23:20[Saint]And if its unreported in any official capacity, as far as I'm concerned it doesn't exist at all.
23:23:49[Saint]WHen you say "this is known", without adding who it is known *to*, can you accept that it is incredibly misleading?
23:25:13[Saint]A user may mistake you for a project member that is actually stating this as fact.
23:25:29chrisjjIf you were misleads, I apologise. Just ignore it.
23:25:41copperit's just a Dothraki saying
23:26:59chrisjjtoehser1: I'm interested in your suggestion about thread safety. I think thread safety is only likely an issue if two or more threads are rendering fonts. Are you suggesting two or more threads are rendering fonts?
23:28:58JdGordonkugel: cool, screenshot?
23:29:22toehser1Interesting. It doesn't usually crash right when it "loses the font" - it crashes much later - say, the font auto-switches, then it works fine but with the default font, then if I pick another theme, THEN it crashes.
23:30:19[Saint]that smells of a buffer handle issue.
23:30:47 Join Scromple [0] (~Simon@
23:31:01[Saint]toehser1: sorry man, I just came into this - do you have a repeatable recipe for a fail case I could look at?
23:31:14[Saint]Or does it "just happen" occasionally.
23:31:24chrisjjInteresting indeed. How about if that other theme has only non-AA font?
23:31:26toehser1So far I can't make it fail in the simulator, but I can reliably make it fail on the player.
23:31:38kugelthere seems to be a problem when a larger font is selected though
23:31:39 Quit Rower (Quit: Hmmm...)
23:31:42toehser1Seems to only happen with certain playlists/songs
23:31:49toehser1and only certain themes.
23:31:54*[Saint] gives kugel the slow clap
23:32:14chrisjjSaint, you might want to know this is the issue of which you said "far as I'm concerned it doesn't exist at all".
23:32:46gevaertschrisjj: fix your attitude *now*
23:32:50toehser1It happens with a theme that uses 13-SourceSansPro-Semibold only, which isn't specially "large" - that font fits 8 lines plus status line.
23:32:53[Saint]Until it is reported in an official capacity, it *doesn't* exist, chrisjj
23:33:02[Saint]So, no, I couldn;t care less.
23:33:29chrisjjOK Saint, but I hope its OK for us to discuss this non-existent issue here.
23:33:29toehser1I don't care if it exists, and I'll eventually report it in an official capacity - but perhaps I'll fix it first...
23:33:30kugelJdGordon: if I select a larger font the 9patch image is broken up
23:33:43Mode"#rockbox +o gevaerts" by ChanServ (ChanServ@services.)
23:33:49Mode"#rockbox +q chrisjj!*@*" by gevaerts (~fg@rockbox/developer/gevaerts)
23:34:17toehser1Sometimes it crashes right in the playing, sometimes when I turn it off, sometimes when I later change themes.
23:34:56[Saint]toehser1: never associated with USB at all?
23:34:56gevaertschrisjj: you're muted now. Learn not to behave like a spoiled eight year old, and I'll unmute you
23:35:11toehser1I'll do the find_addr when I get the chance. (Right now, I'm hobbled in the the player doesn't have the build that I have access to over vnc/ssh from git, but when I get home, I'll make the player run the build I have in source...)
23:35:30toehser1Doesn't seem to matter. I can unplug from USB, reboot, and make it fail standalone.
23:35:35JdGordonkugel: screenshot of the broken one?
23:36:23[Saint]toehser1: And this is repeatable with git head?
23:36:30 Join Rower [0] (
23:36:43 Quit Rower (Client Quit)
23:36:44[Saint]If you have a recipe, even a vague one, I can try to repeat it and see if I can catch it.
23:36:56[Saint]This seems a very interesting issue.
23:37:05toehser1I'll be ensuring that within a couple of hours when I'm home.
23:37:07[Saint]And I would very much like it confirmed.
23:37:19[Saint]Aha. Great. Thanks.
23:37:51toehser1The recipe seems to be: Clip Zip, one of the themes I created (tomsway1/2/3), and "certain" files/playlists... and it works for a while before failing, not right away fail.
23:38:11copperforums down?
23:38:36 Join Rower [0] (
23:38:36 Quit Rower (Client Quit)
23:39:02AlexPcopper: Seem to be here
23:39:13AlexPThey were working a minute ago though
23:39:15Mode"#rockbox +o AlexP" by ChanServ (ChanServ@services.)
23:39:19toehser1When I get home in a couple hours, I'll update git, make sure it fails, do the find-addr, etc. etc.
23:39:33gevaertscopper: yes and no. My guess is that someone is being banned, and the database needs a few more indices. scorche|sh should gave some time soon to add those, and I hope things will be better then
23:39:45kugelJdGordon: what happens if the sizes if the middle segments is not a multiple of a third of the patch image?
23:39:52[Saint]copper: AlexP: - I just got through, but it took ages.
23:40:09AlexPIt'll be what gevaerts said I think
23:40:15kugeli think that might be the problem
23:40:19AlexPBanning someone kills the forums at the moment
23:40:27[Saint]kugel: from memory it does as little overlapping as it needs to, decided by the draw ordering.
23:40:28JdGordonkugel: i dont remember the patch at all :)
23:40:38JdGordonand im off to work
23:41:11[Saint]back up now.
23:41:34[Saint]So, yeah, probably someone poking at the db, which seems to grind things to a halt.
23:44:40Mode"#rockbox -q chrisjj!*@*" by gevaerts (~fg@rockbox/developer/gevaerts)
23:44:58AlexPGah, who is banning now!
23:45:22gevaertschrisjj: ok, you're unmuted again, but try not to annoy everyone again
23:46:09copperI never got through
23:46:57chrisjjtoehser1: [repeat of blocked submission of 22:35] "right in the playing" sounds like the case to focus on. Is this right in the WPS without going to any other page?
23:47:40chrisjjClarification: can you gent the font loss, and then without leaving the WPS, get the crash?
23:49:34toehser1When it fails that way, yes.
23:49:44copperah, got it now
23:49:54toehser1Um... It may just directly crash with out even the font loss... I'll check.
23:52:12toehser1What happened just now was the font change...
23:52:17toehser1And it hasn't crashed yet.
23:53:12toehser1There has to be code doing this font change, something like "if error with X, switch to default font, continue", trying to recover from something - hard for me to think it would switch fonts without something like that.
23:53:22 Join Rower [0] (
23:55:04toehser1Didn't fail, kept working with wrong font. But then, I did power-button to turn off... and "Data abort at 30 FSR 0x8 ...
23:55:08[Saint]I don't believe any such code exists. Once the font is loaded, it shouldn't ever need to be freed from memory.
23:55:22[Saint](unless it is unloaded)
23:56:27[Saint]But I will admit that buflib has many minutiae I am unaware of.
23:56:32toehser1well... the behavior is, theme works, works, works, suddenly font changes, works, works, crash...
23:57:03[Saint]But I know that (well, I _think_ I know that) nothing should be asking the skin system to free fonts that are in use.
23:57:33toehser1I'll have to read code, and research things, before I talk more on it... at the moment, I'm hobbled by my work computer environment, which allows me to VNC and SSH to my home computer, but which does _NOT_ allow me to connect a USB device...
23:57:40PurlingNayukiJdGordon: Yes i I was sleeping
23:57:51[Saint]When the font changes, does it change to an arbitrary user font (one the skin is also using elsewhere?), or the sysfont (ugly, fixed width tiny 8pt font)
23:58:04toehser1Always sysfont
23:58:27toehser1The theme only uses 1 font, set in the .cfg, and referred to with ,1
23:58:39kugel[Saint]: what draw ordering?
23:58:54 Quit lebellium (Quit: ChatZilla [Firefox 27.0/20131216183647])
23:58:54toehser1The WPS doesn't do anything but use ,1 for all the viewports

Previous day | Next day