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 2011-04-21

00:00:34bluebrotherwell, I'm seeing the "Original Remote Commands" according to this website. Well, will check that tomorrow.
00:04:44 Quit avacore (Ping timeout: 246 seconds)
00:06:49 Join avacore [0] (
00:08:53 Join scromple [0] (
00:11:00 Quit kevku (Ping timeout: 260 seconds)
00:12:41bertrikwhat is the "G3 protocol"?
00:13:48 Join sideral [0] (~sideral@
00:13:49 Quit sideral (Changing host)
00:13:49 Join sideral [0] (~sideral@rockbox/developer/sideral)
00:17:05saratogai think its the original protocol on the 3G ipods
00:25:08 Quit bertrik (Quit: :tiuQ)
00:33:03 Quit Unhelpful (Quit: - Chat comfortably. Anywhere.)
00:47:43***Saving seen data "./dancer.seen"
00:55:24 Quit silbo (Remote host closed the connection)
00:56:15 Quit wodz (Quit: Leaving)
01:00:26 Quit milk (Ping timeout: 258 seconds)
01:13:30 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
01:14:40 Part toffe82
01:18:22 Join Keripo [0] (
01:18:24 Quit bluebrother (Disconnected by services)
01:18:25 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
01:18:26 Quit sideral (Quit: Leaving.)
01:23:16 Quit TheLemonMan (Quit: Destructor called)
01:33:08 Join T44 [0] (
01:34:18 Quit Topy44 (Read error: Connection reset by peer)
01:35:29 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
01:37:48 Quit robin0800 (Remote host closed the connection)
01:54:34 Quit bieber_ (Remote host closed the connection)
02:00:40 Join Strath [0] (
02:07:31 Join L-Strife89 [0] (~Strife89@
02:08:59 Join reacocard [0] (
02:10:29 Quit Strife89 (Quit: Reboot.)
02:10:37reacocardthe microsd in my e200 isn't showing up on the computer (even via fdisk -l or dmesg) when i plug it in over usb. I'm using rockbox 3.8.1 and arch linux, any ideas?
02:10:51reacocardit shows up fine in rockbox on the player itself
02:19:13 Join bieber_ [0] (
02:23:07reacocard..apparently its only an issue with my netbook, a different computer reads it fine. go figure.
02:23:14 Part reacocard ("Sleep peacefully, my most beloved witch, Beatrice.")
02:35:04 Join WilliamC [0] (~William@
02:35:14WilliamCHow do I upload my Scrobbler logs?
02:35:31WilliamCThe website I usually use seems to have somewhat crashed
02:41:59 Join JoshuaChang [0] (
02:47:45***Saving seen data "./dancer.seen"
02:48:00 Quit JoshuaChang (Quit: CGI:IRC (EOF))
02:53:29 Part WilliamC ("Leaving")
02:54:33 Join jo__ [0] (~JoshuaCha@
02:55:54 Nick jo__ is now known as JoshuaChang (~JoshuaCha@
03:07:52 Quit JoshuaChang (Quit: ChatZilla [Firefox 4.0.2pre/20110415144045])
03:22:19 Join JoshuaChang [0] (~JoshuaCha@
03:52:54 Join milk [0] (
03:57:22 Join Leif [0] (
03:59:53 Join jaws [0] (~fake@
04:02:15jawswas wondering if anyone can add some info into the wiki? I think this is important for the voicing of file names ETC . a full list of codes is found at this link and you would enter the language ids into the script without the leading 0, since peole have trouble with scansoft voices
04:02:39JdGordongo for it
04:02:44jawsso for instance the spanish voices use the language id 41A instead of the one that is listed in the site 041A
04:02:44JdGordonits a wiki after all
04:02:51jawsthe link is..
04:02:57jawsI am not registered though
04:03:15JdGordonso register
04:03:37jawslol. though that would be the only thing I would have to contriute for now,
04:17:24 Quit tchan (Quit: WeeChat 0.3.4)
04:35:12 Quit simonlnu (Quit: brb)
04:37:49 Quit amiconn (Disconnected by services)
04:37:51 Quit pixelma (Disconnected by services)
04:37:51 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:37:53 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:37:55 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:38:08 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:41:47 Quit Leif (Ping timeout: 260 seconds)
04:47:48***Saving seen data "./dancer.seen"
04:51:25 Quit TheSeven (Ping timeout: 276 seconds)
04:51:41 Join simonlnu [0] (simon@unaffiliated/simonrvn)
04:53:44 Join Leif [0] (
04:53:46 Quit milk (Quit: baaaiiii)
04:54:28 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:58:54 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
05:02:12 Quit kugel (Ping timeout: 260 seconds)
05:16:18 Join Rob2223 [0] (
05:19:56 Quit Rob2222 (Ping timeout: 240 seconds)
05:27:30 Join liuft [0] (~liuft@
05:41:52 Quit Horscht (Quit: Verlassend)
05:47:00 Quit Strath (Quit: A day without sunshine is like .... night)
05:48:21jhMikeSI'm done messing around with update (except for basic sync) for now until I may dump this load:
05:53:20JdGordonjhMikeS: do a set of test builds?
05:54:24jhMikeSFine, which are most important to do?
05:58:29jhMikeSof course, dump it in SVN and all targets suddenly become test builds :)
06:01:29 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
06:04:19JdGordonthats the easier option sure :)
06:04:39JdGordondo a full set and wait a few days for the amazing lack of enthusiasm
06:06:19 Quit ChickeNES (Ping timeout: 248 seconds)
06:06:29jhMikeSexactly my point, stirring the pot is much more effective in my experience
06:07:28JdGordonyeah, but it looks better at least trying to get testers :p
06:08:02saratogai've been playing with it and havne't had any issues, aside from that SPC thing which happens with SVN anyway
06:09:02jhMikeSsaratoga: how do you manage that? the sorts of actions you talk about are pretty much what I carry to the extreme before submitting a patch in the first place.
06:09:21LloreanHow long do we have until the next release? Wouldn't it be better to get it in SVN now?
06:09:27saratogajust picked any SPC file and jam the select on the 200 as fast as I can
06:09:44jhMikeSsaratoga: me too, nothing comes of it other than it repeats
06:10:23saratogacrashes everytime for me, let me check my config file
06:10:23jhMikeSLlorean: I think a couple of months ??
06:10:53jhMikeSsaratoga: feel free to send me that if you don't mind (there's nothing special about starting SPC over any other codec really)
06:11:02LloreanjhMikeS: Yeah, that's what I was thinking. You don't want to sit waiting for testers too long, I'd say if you think it's working, go for SVN. We have a relatively recent release if it is unexpectedly buggy, and we get a very quick test response.
06:11:24jhMikeSLlorean: been using the basic changes for awhile now, exclusively
06:11:51jhMikeSLlorean: right, quick revert if it's a complete catastrophe :p
06:12:04saratogaheh clearing the cfg file fixes it, although the config is unremarkable
06:12:09saratogaexcept maybe for repeat all
06:14:30saratogatry jamming select a couple times and then selecting a different track
06:14:35saratogathat also seems to freeze for me
06:15:45saratogamy config if it matters:
06:16:00 Quit tchan (Quit: WeeChat 0.3.4)
06:16:05LloreanjhMikeS: I'm honestly all for using SVN to test things on users as long as there's time to get a meaningful response, and revert it, before the next freeze in case anything unexpected goes wrong
06:16:13LloreanI'd say both of those are probably about as true as they get right now
06:17:17saratogajhMikeS: yeah repeat all needs to be on
06:17:37saratogaso its a playback.c bug, probably just happens in spc due to SPC files being so small
06:17:42jhMikeSThen I'll happy mash the commit button on it and monitor
06:18:27jhMikeSsaratoga: maybe. it might be because playlists against playback is quite racy
06:18:38*jhMikeS gives it a go
06:21:18jhMikeSyou've got other options like replaygain
06:21:21L-Strife89What port does svn try to use by default?
06:23:06saratogai cleared the settings, then enabled repeat all, and that was enough to make it happen
06:23:25saratogajamming one track a couple times, then trying to play another track seemed the easiest way to reproduce it
06:23:41jhMikeSI hit it with dircache on and repeat all on
06:23:56saratogadircache is on by default i think, so i probably had that too
06:24:48jhMikeSno dircache, no problem, repeat one also works
06:25:09saratogawithout dircache the timing is quite different IIRC
06:25:31saratogaplayback.c handles metadata somewhat differently, and the wps updates differently i think
06:25:54jhMikeSplayback.c knows no difference
06:26:09jhMikeSnot internally anyway
06:27:18JoshuaChangwell, seems the repeat all setting is the key?
06:27:29jhMikeSall/one + dircache
06:27:45saratogasomewhere in the guts of playback it affects something, i remember running into this a while ago
06:27:52saratogamaybe not playback.c though
06:28:26jhMikeSplayback is going to whizz through the files very quickly and if it starts altering the list while playback is reading it then it may very will have issues
06:28:35JoshuaChangrepeat all means to read all the tags while buffering?
06:29:10saratogafor really tiny files like SPC, does repeat all mean its going to buffer each file a lot of times?
06:29:24JoshuaChangin simulator, i can see the console report a lot of tag reading action while select a file
06:29:27jhMikeSrepeat all just repeats the current directory over and over, repeat one, the current song over and over but playback just sees what playlist is giving it
06:29:53saratogaso it doesn't buffer past the last track?
06:29:58jhMikeSsaratoga: indeed it will, the playlist won't have an end
06:30:24jhMikeSsaratoga: no playlist just gives it the first file again after the last one, which repeat one, it gets the same file over and over
06:30:53jhMikeSit really doesn't know anything about what playlist is doing
06:31:14JoshuaChangi still think the bug is related to buffering and tag reading, maybe they bump together in the cpu/memory?
06:31:17saratogaso its buffering each one of those tiny SPC files about a thousand times
06:31:46jhMikeSsaratoga: it will do that yes
06:31:59saratogathat certainly gives it a rather large window for a race condition
06:32:39jhMikeSput it on repeat one and it should fill the whole track array
06:32:44jhMikeSall with the same file
06:34:03JoshuaChangwhile in repeat all mode, you select one file, and the rb will read all the files' tag in the same dir
06:34:21saratogait does that regardless of repeat all
06:34:37JoshuaChangin repeat one mode, rb only read the file tag you selected
06:34:44jhMikeSyes, it should do that, but it will restart at the first one in repeat all when normally the playlist would end at the last song
06:35:11JoshuaChangi can see that in the simulator's console
06:35:14jhMikeSright, in repeat one, it buffering the same track as many times as it can
06:36:01JoshuaChangand the buffer is bumped by the file you are current playing
06:36:41jhMikeSI wonder if shuffle does it too
06:37:22JoshuaChangshuffle also need all the files' tag
06:37:50jhMikeSit always needs all the files' tags or playback can't do anything with it
06:38:21jhMikeSno luck with shuffle
06:38:33JoshuaChangin one mode, it only need current file's tag
06:38:41jhMikeSA-B repeat, same thing
06:39:00saratogathose won't setup infinite playlists, so thats probably why
06:39:14*jhMikeS meant A-B is just like one/all
06:39:41saratogait crashes?
06:39:58saratogadoes A-B repeat just buffer the same track over and over
06:40:03saratogai assumed it skipped backwards
06:40:09jhMikeSyes, a-b gives the same song just like repeat one
06:40:40jhMikeSa-b is the essentially the same as repeat one + you can set your markers
06:41:47jhMikeSit does seeks but if no b-marker is set, it does the "next" track, which is the same one again
06:41:48JoshuaChanga-b only read current file, simulator console reported
06:42:26saratogaah so its basically the same as repeat one
06:42:32saratogamakes sense then
06:42:39JdGordonjhMikeS: is there any point in putting the magic logic into playback to not rebuffer a track on repeat-one/a-b if the whole track fits in the buffer?
06:43:31jhMikeSplaylists just clobber the current playlist while buffering is happening on that list
06:44:19jhMikeSJdGordon: it wouldn't rebuffer anything if the whole track fits, unless you mean "repeat it".
06:47:52***Saving seen data "./dancer.seen"
06:49:53 Join kevku [0] (~kevku@2001:470:28:773:babe:feed:dead:bee)
06:50:16jhMikeSI did spot something odd where I'm only getting 127 tracks buffer but it should be 128 (but doesn't appear to have any negative effects)
06:52:56 Join sideral [0] (~sideral@rockbox/developer/sideral)
06:57:35 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
07:00:28jhMikeSwhere does the click get picked up, filetree?
07:06:01jhMikeSthe FILE_ATTR_AUDIO case in filetree
07:07:11 Quit sideral (Quit: Leaving.)
07:07:13jhMikeSplayback isn't going to know about any of that happening for quite a long time
07:08:23JdGordonthaht goes through ft_enter and then playlist
07:11:49 Quit Leif (Quit: Leaving)
07:13:06jhMikeSit's reading the playlist really quick but doesn't know it's different yet since Q_AUDIO_FLUSH is seen whenever it's seen, but who know when that will really happen
07:14:02jhMikeSor in this case, Q_AUDIO_PLAY
07:22:54jhMikeSI was using a current SVN build on e200, getting the lockups (which explains the funny track and handle count) :)
07:23:27jhMikeSthought I put a patched one on there after that :P
07:24:07jhMikeSsaratoga: test_codec was going ok?, it's built in that patch
07:24:34saratogai didn't try
07:25:25saratogaseems to be running
07:25:53saratogayeah seems fine
07:30:56 Join Strife89 [0] (~Strife89@
07:35:43 Join ChickeNES [0] (~ChickeNES@
07:44:16jhMikeS+! :)
07:44:49 Join Buschel [0] (
07:47:01jhMikeSI think this playlist issue is out of the scope of what I intended to work on, which was basically control-related stuff
07:52:27 Quit L-Strife89 (Quit: Bed.)
07:56:56jhMikeSThe only real easy, do-it-for-now fix is to stop playback before modifying the playlist, but then pretty crossfades when you select new tracks won't work
07:57:20JoshuaChangi want to add some audio effect, but i can't fully understand the dsp.c 's processing mode, what's meaning of int32_t *buf[] in some function?
07:57:30 Quit GodEater (Remote host closed the connection)
07:57:50JoshuaChangdoes the int32_t *buf[] refresh every sample?
07:57:51jhMikeSpointers to each channel's (L and R) data
07:58:36jhMikeSno, just every chunk, which varies depending on how it slices it up
07:59:43jhMikeSinternally everything is non-interleaved
08:00:10JoshuaChangyou mean, if set the buf[] to 200, then the function eat 200 samples every time and process them, then push them all?
08:00:24BuscheljhMikeS: so, when do you think you might submit your playback engine rework?
08:01:22jhMikeSsoon, when I feel a comfortable enough vibe from enough people
08:02:00saratogaJoshuaChang: buf[] is an array of samples, not a scalar
08:02:26*Buschel sends out positive vibes
08:02:35jhMikeSJoshuaChang: buf[0] is pointer to left data, buf[1] is pointer to rightt data (if it's not mono). those increment by the number of samples processed in each step
08:02:49jhMikeSBuschel: ommmmmmmmmm.....
08:03:38JoshuaChangehh, my poor english...
08:04:30JoshuaChangwell, if define buf[200], then in stereo mode, it should carry 100 samples, right?
08:04:43BuscheljhMikeS: do you think there is a chance of FS #11495 being fixed as well? nasty bug :/
08:08:43jhMikeSBuschel: hmmm...I'll go look
08:09:08JoshuaChangi need a delay process, which means the output samples may accumulate with the samples 0.5 second ago, can buf[] do this job?
08:10:14jhMikeSbuf[] exists in the caller's space most of the time, I'd suggest a delay line of some sort
08:10:47 Quit amiconn (Remote host closed the connection)
08:10:49 Quit pixelma (Remote host closed the connection)
08:11:44JoshuaChangshould i create my own buffer and do the delay line function myself?
08:11:49jhMikeSBuschel: all I need is a atrac3_rm file. the core won't set the codec's current position anymore for it, so it might get done correctly now
08:11:59jhMikeSJoshuaChang: yep
08:12:30JoshuaChang……that’s too hard for a newbie like me 他哦
08:12:36JoshuaChangto do...
08:13:19*jhMikeS was a newbie but got obsessed
08:14:00jhMikeSand still feels like a newbie most of the time
08:14:02BuscheljhMikeS: can you access this link to download rm-samples?
08:14:51jhMikeScan't seem to
08:15:22Buschelcan you read 7zip?
08:16:13jhMikeSexcept the ones [Saint] sends for some reason, even though I use the official program
08:16:30BuschelI'll upload some files for you, give me a few minutes
08:17:14 Join leavittx [0] (~leavittx@
08:17:36*jhMikeS should make note that the newer code doesn't display UI from the audio thread, it just skips
08:18:42*jhMikeS notices he has a sample file there
08:19:45 Join amiconn [0] (quassel@rockbox/developer/amiconn)
08:19:48 Join pixelma [0] (quassel@rockbox/staff/pixelma)
08:20:47saratogajhMikeS, Buschel:
08:20:47jhMikeSoops, it's at the .hu url
08:22:34jhMikeSI haven't had problems with cook and the codec is similar in style (same author)
08:26:41 Join ender` [0] (
08:26:54jhMikeSit seems to be working as far as hanving anything playing else first. it's only resuming at 0 though :\
08:28:22jhMikeSsomething interesting did happen though, it was playing along nicely for awhile and then started just buzzing
08:29:49 Join sideral [0] (~sideral@rockbox/developer/sideral)
08:31:42jhMikeShmm...atrac3 already seeks the buffer beforehand in svn
08:32:42Buschelgotta go to work now
08:32:44Buschelsee you
08:32:47 Quit Buschel (Quit: ChatZilla [Firefox 3.6.16/20110319135224])
08:33:03jhMikeSoops, had that tree patched
08:37:16jhMikeSheh, forgot a test to initially skip the comand polling upon resume
08:46:59 Join bertrik [0] (
08:46:59 Quit bertrik (Changing host)
08:46:59 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:47:55***Saving seen data "./dancer.seen"
08:49:16 Quit jaws (Read error: Connection reset by peer)
08:52:06 Join n1s [0] (
08:52:06 Quit n1s (Changing host)
08:52:06 Join n1s [0] (~n1s@rockbox/developer/n1s)
08:52:15 Join mem_ [0] (
08:52:16jhMikeSsvn still bombs in the same place on DearJohn
08:54:30 Quit leavittx (Ping timeout: 276 seconds)
09:00:10jhMikeSBuschel: (for logs) one thing the patch guarantees is the codec has the correct metadata at the time codec_run is called (but _not_ in codec_main, where the id3 is undefined)
09:00:32 Join LinusN [0] (
09:04:23 Quit scromple (Quit: Gots to go get me outta here)
09:09:04 Quit advcomp2019 (Read error: Connection reset by peer)
09:09:20 Join leavittx [0] (~leavittx@
09:09:53 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
09:14:30 Join DerPapst [0] (
09:23:19sideraljhMikeS: Re playback rewrite: IIUC, you're saying the patch could be split. It would be great if you could post a patch set with several logically distinct and self-contained patches −− that would simplify review quite a bit
09:24:28jhMikeSBuschel: everything about playing seems on par with SVN and I haven't had any failure from selecting a file
09:25:18 Join [Saint] [0] (~st.lasciv@
09:25:31 Quit [Saint] (Disconnected by services)
09:25:33 Join S_a_i_n_t [0] (~st.lasciv@
09:26:03jhMikeSsideral: Only the buffering and kernel bits really, which do nothing to speak of for current SVN
09:26:42 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
09:26:50jhMikeSthe rest just doesn't operate quite the same way between codecs and metadata handling
09:27:06sideralI see. Splitting the simple parts off would probably still be helpful
09:27:43gevaertsjhMikeS: I don't think people are asking for separate *useful* patches.
09:28:19gevaertsThey want patches that are small enough to review and contain more or less standalone changes, even if those changes make no sense without the full set
09:28:40jhMikeSgevaerts: well, I could commit the those extensions now and not touch playback, mostly they'll just add idle binsize
09:28:49 Quit bertrik (Quit: :tiuQ)
09:29:13gevaertsPatches, not commits :)
09:30:11jhMikeSanyway, lots of codecs, mostly minor updates in each
09:30:19jhMikeSit makes it look very big
09:31:41 Quit leavittx (Read error: Operation timed out)
09:31:56jhMikeSok, I'll split it up (lazyhead) :p ...
09:32:22 Quit S_a_i_n_t (Disconnected by services)
09:32:24 Join [Saint] [0] (~st.lasciv@
09:32:59 Join leavittx [0] (~leavittx@
09:33:07sideraljhMikeS: BTW, Re AMSv2 USB corruption: I've played with the USB driver a bit and (1) noticed that sync differs to the AMSv1 driver in one regard, and (2) I found one a case of data corruption due to the SETUP packet buffer that's apparently too small. But unfortunately fixing these issues did not fix the problem for me
09:33:48sideralI find the SETUP handling rather strange, especially the "ignore spurious setup" bit
09:34:31jhMikeShmmm...would sync be expected to be the same? it's different hw isn't it/
09:35:05sideraljhMikeS: I meant sync with driver callers, not sync with hardware
09:36:03jhMikeSand if a buffer is too small, it's probably best to just fix that
09:36:05sideralThe difference is that the AMSv1 driver calls usb_core_transfer_complete only when the xfer call was nonblocking, whereas the AMSv2 driver calls it always
09:36:41sideralbut that apparently doesn't hurt too much, as there will be no handler for it in the blocking case
09:37:31 Quit shai (Ping timeout: 260 seconds)
09:37:44sideralI wanted to discuss the SETUP buffer change with pamaury first, who apparently has an actual idea of what the hardware does :)
09:38:28 Quit ChickeNES (Quit: Computer has gone to sleep.)
09:38:29jhMikeSyeah, probably a good idea :) I'd also like to know where some extra bit defines came from since I didn't find them in the datasheet that was supposedly used.
09:39:53sideralHe's pointed me to the linux-as353x patch we have access to, but maybe I mix things up and that was for something else
09:43:24 Quit [Saint] (Disconnected by services)
09:43:26 Join S_a_i_n_t [0] (~st.lasciv@
09:49:51sideraljhMikeS: That Linux driver references the databook for a Synopsis DWC OTG, which seems to be the USB IP block that was used. Possibly related to , which offers a Linux driver (registration required)
09:55:26 Quit JoshuaChang (Remote host closed the connection)
09:58:31 Nick kugel_ is now known as kugel (~kugel@rockbox/developer/kugel)
10:04:56 Join smk [0] (3ba3df59@gateway/web/freenode/ip.
10:12:32smkhi anybody here who knows how to run rockbox simulator with valgrind?
10:13:45 Quit Rob2223 (Quit: Rob2223)
10:13:50gevaertssmk: the first thing to do for issues when running the simulator with debugging tools is building with sdl threads
10:14:17gevaertsi.e. pass −−sdl-threads to tools/configure
10:14:26 Join Rob2222 [0] (
10:14:54gevaertsAnd probably make clean, this is a big enough change that I suspect the build system will get confused otherwise
10:14:57smkoh. yeah. it hangs in the setup_threads function called by SDL.
10:15:57gevaertsThe default threading system is faster for our use and closer to what's used on real hardware, but it does things that debuggers don't like
10:18:12smkwill rebuild then.
10:20:40gevaertsWell, I'm assuming you're not actually interested in debugging the threading code of course :)
10:20:41 Join pamaury [0] (81680b01@rockbox/developer/pamaury)
10:21:30smkno. i am having some memory related problems. made some changes (replaced some malloc calls with static allocations) and its causing weird and funny errors :)
10:21:31kugelgevaerts: the sigaltstack threads work fine just with gdb
10:21:42kugel(valgrind has indeed problems though)
10:22:33gevaertskugel: yes, provided you set up the right things. I know that if I need gdb, and the problem isn't related to threads, I use −−sdl-threads because it's less hassle
10:24:02kugelwhat do you mean by hassle? do you mean that it triggers on SIGURS1 on start up?
10:24:13*gevaerts nods
10:24:19gevaertsIt's one line more to type!
10:24:46kugelIIRC you can put "handle SIGUSR1 pass nostop" into some config file so that's it always executed
10:24:49sideralpamaury: Re AMSv2 USB corruption: I've played with the USB driver a bit last night found one a case of data corruption due to the SETUP packet buffer that's apparently too small. But unfortunately fixing it did not fix the corruption problem for me
10:25:30gevaertskugel: yes, but that assumes I'm only ever using gdb with the sim
10:25:38sideralpamaury: Can you explain the "ignore spurious setup" bit in the driver?
10:26:08kugelyou could also setup a rbgdb which does that
10:26:39kugelOTOH you make sure sdl threads keep working, not the worst thing :p
10:26:48gevaertsexactly :)
10:26:56*gevaerts knew he wasn't *just* being lazy!
10:27:22 Quit pamaury (Ping timeout: 252 seconds)
10:27:50 Quit S_a_i_n_t (Quit: I'm only going to Heaven if it feels like Hell, I'm only going to Heaven if it tastes like caramel...)
10:29:53 Join [Saint] [0] (~St.]@
10:30:38jhMikeSgevaerts: :>
10:43:35sideralThank you jhMikeS!
10:44:45jhMikeSyour welcome!
10:45:17jhMikeSI see I left an experimental thread_set_priority in buffering.c :\
10:47:57***Saving seen data "./dancer.seen"
10:56:07 Join ricemark20 [0] (
10:58:25 Join TheLemonMan [0] (
10:58:43 Part smk
11:03:22[Saint]JdGordon: How does one alter the size of the startup logo splash?
11:04:10CIA-87New commit by kugel (r29753): Android: Fix creation of debug key if ~/.android doesn't exist.
11:06:13[Saint]Oh, it wasn't just me that KEYTOOL was being annoying for I see.
11:06:35[Saint]Always good when you find out something's not your fault. ;)
11:07:15CIA-87r29753 build result: All green
11:21:14 Quit liuft (Quit: Leaving)
11:22:10 Quit DerPapst (Quit: Leaving.)
11:28:52 Quit ricemark20 (Quit: Leaving)
11:29:07 Quit lennyk (Quit: lennyk)
11:30:45sideralkugel: Your last commit should probably use "mkdir -p" unless you want it to fail of the directory already exists
11:32:38kugelsideral: it doesn't error out for me
11:33:11gevaertsIt should
11:33:36sideralthen either you have an alias defined (many systems have a stock alias mkdir mkdir -p) or you're not on a POSIX box :)
11:33:50kugelah, of course. it errors if .android exists but not debug.keystore
11:33:54gevaertssideral: I wouldn't expect make to use aliases
11:34:04kugelif debug.keystore also exists the make target isn't made
11:34:49sideralgevaerts: I just guessed kugel was trying it out in the shell ;)
11:35:21kugelI tried with and without .android, but not with an empty .android :)
11:35:38CIA-87New commit by kugel (r29754): Use mkdir -p in the previous commit. Thanks sideral for spotting.
11:38:45CIA-87r29754 build result: All green
11:54:05*jhMikeS gets crazy ideas to facilitate supporting "exotic" formats in playback, like midi and hybrid wavpack
11:56:04 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
11:59:57jhMikeSbuffering should just avoid dealing with files directly by default and use some reader source so it doesn't care where or how it gets the data to buffer
12:18:49 Quit saratoga (Ping timeout: 252 seconds)
12:40:33JdGordon[Saint]: err.. it should be compiled in, i dont think the code cares about its dimensions
12:43:56 Quit kevku (Quit: KVIrc 4.0.4 Insomnia
12:44:10 Quit rdd (Read error: Connection reset by peer)
12:44:36 Join skapazzo [0] (~skapazzo@
12:45:12 Join pamaury [0] (81680b01@rockbox/developer/pamaury)
12:45:41pamaurysideral: ping
12:46:01pamaurysorry for missing your poke, I was using the web interface of freenode and it silently disconnected :-/
12:48:00***Saving seen data "./dancer.seen"
12:48:42pamauryI can't how can the setup buffer be too small, a setup packet is always 8 bytes. The only corruption case *would* be that the host sends another setup packet without waiting for the previous one to complete, I think there is a mechnism against it but I've never seen anything like this ever happen
12:50:41pamaurythe spurious test is in case the driver receives a setup packet of invalid length (ie !=8). I had some of those in the early development, but it should not happen
13:09:12sideralpamaury: I frequently see this, and in one case I could see that the variable allocated behind the packet buffer (ep0_state) was corrupted
13:09:36pamauryis this in the case where the controller keeps resetting ?
13:10:36sideralPerhaps the size reported by the "ignore spurious setup" message corresponds to bytes written into memory after the 8-byte buffer?
13:10:36pamaurythe corruption makes no sense, the dma is configured to transfer 8 bytes, not more !
13:10:57sideralI don't know whether you can really limit the xfer size that way
13:11:13sideralThis is not done the same way for non-EP0 non-SETUP packets
13:11:24pamauryyes you can, the reported size is the amount missing from the 8 bytes
13:11:32sideral(by writing to the xfersize field)
13:11:40pamauryhum ? There is no difference
13:12:17sideralah yeah, right
13:12:32 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:12:47 Quit ender| (Ping timeout: 250 seconds)
13:13:13sideralI don't think this was the permenent-reset case. The USB connection eventually just hung
13:13:13pamauryyou intiially set DOEPTSIZ to the maximum size it gets decrementented. The controller fires the irq either on short packet or when the count reaches zero AFAIK
13:13:49 Quit ender` (Ping timeout: 246 seconds)
13:14:10sideralSou you'd expect the size reported in the "ignored" message should be < 8?
13:14:31pamauryit can't be more
13:15:00pamaurywell, it shouldn't be since the controller can only decrement it and I set it to 8 !!
13:15:02sideralI observed values of xfersize=120 and xfersize=112
13:15:50pamauryon ignore spurious setup ?
13:16:04sideralbut it's possible that I made a mistake −− I experimented with not setting the xfersize bits
13:16:38pamauryarf, not setiing the xfersize bits seems a bit dangerous and weird
13:17:05sideralit could have been that experiment. (I really should better log my experiments :/ )
13:18:16pamauryI won't have time until tomorrow night but since my device behaves differently without sd, I'll try to understand what is the difference between sd inserted and sd not inserted
13:18:46sideralcould it be a problem that we set both DEPTSIZ0_supcnt_bitp and DEPTSIZ0_pkcnt_bitp?
13:19:19 Quit otih (Remote host closed the connection)
13:19:26pamauryI honestly can't remember what those bit do :) And I don't have any document at hand, I'll have a look at home
13:20:18sideralOK. :) One other bit of difference I found between the AMSv1 and AMSv2 USB drivers was when they send usb_core_transfer_complete
13:20:20pamauryiirc, the pkcnt one is the the number of packets to received (at most) and supcnt is the number of setup (for a control endpoint)
13:20:56sideralHopefully that can't lead it to receive two packets (one of each type)?
13:21:34pamauryno, afaik the only part using blocking transfers is the core and it ignores transfer completion
13:22:07sideralyeah, changing that to the AMSv1 way did not make a difference
13:23:01pamauryI would even say the AMsv1 is unsual then because the usb-arc driver also behaves like this: always call usb_core_transfer_complete
13:23:18pamauryand usb works reliably on all arc-based devices
13:23:53sideralMisunderstanding? My "receive two packets" line referred to DEPTSIZ0_supcnt_bitp and DEPTSIZ0_pkcnt_bitp being set
13:24:50 Quit Rob2222 (Read error: Connection reset by peer)
13:24:54sideralWhat kind of documentation do you have? Just the linux-as353x patch?
13:25:02pamauryI was thinking about that one but I dont't think so, although one should check. and in any case, the xfer size is decremented so the second one would have 0 size
13:25:04 Join Rob2222 [0] (
13:25:40pamaurysideral: that patch but it's not extremely useful, I have some datasheet for similar-but-not-exactly-the-same-controller-and-those-datasheet-are-not-complete :)
13:26:33sideralIs the datasheet for the Synopsys USB controller?
13:27:21pamauryno, it's for a SoC and it has a partial documentation for the registers. We don't have an official doc from synopsys
13:27:40sideraltoo bad.
13:27:54pamaurythe patch has lots of doc inside for the registers but that doesn't help
13:28:20pamaurybecause the interesting part is how you init the controller, how you setup a transfer, things to avoid, etc
13:29:08pamauryOn idea is popping out of my mind: I think this controller has a PIO mode so I could try to implement a PIO driver
13:29:12pamaury* one
13:29:29sideralnice idea
13:30:02pamauryit's far less efficient and more complicated to implement I think but at least we know where we write in memory :D
13:34:01sideralOne thing I find confusing about the linux patch is that it appears to have multiple USB drivers
13:38:15 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
13:44:29 Quit skapazzo (Ping timeout: 258 seconds)
13:51:12 Join skapazzo [0] (~skapazzo@
14:01:43 Join robin0800 [0] (
14:05:53pamaurysideral: really ?
14:07:02pamauryIt's much more general: it has the device side and the host side (since it's OTG), and it handles DMA vs PIO (and even internal vs external DMa), it's a very general framework
14:07:15sideralmaybe I was just confused, but I think found two places where the SETUP DMA buffer is set up
14:07:25sideralah, that explains it
14:07:56pamauryActually, half of the files are not of interest because of OTG :)
14:12:26 Quit robin0800 (Quit: Leaving)
14:16:53 Quit T44 (Quit: Leaving)
14:22:05sideralpamaury: Actually, OTG would be pretty neat to have on my Clip+. I could plug in a WiFi stick and sync it over the air :)
14:22:25pamauryyeah but we don't know if it has the electrical wiring ;-/
14:27:29 Join robin0800 [0] (
14:39:33ThomasAHpamaury: is the wiring only relevant for deciding if OTG mode should be enabled, or for other things, too? (assuming 5V can come from a powered USB hub)
14:40:21 Quit pamaury (Ping timeout: 252 seconds)
14:44:22 Join einhirn [0] (
14:45:53sideralThomasAH: I take it you are planning to port a host USB stack, an IP stack and a few WiFi drivers to Rockbox? ;)
14:48:02***Saving seen data "./dancer.seen"
14:50:49 Quit robin0800 (Quit: Leaving)
14:53:25ThomasAHsideral: no, I hoped to use OTG on my Motorola Defy, but it only works for some seconds directly after booting before the normal system takes over :)
14:53:53ThomasAHsideral: and I hope to see a working USB storage for my trusted Clip+, too :))
14:55:26ThomasAHsideral: (so I hoped to ensnarl pamaury into a discussion about Clip+ and USB by throwing in my half-knowledge about USB OTG so I could ask him about the status of USB on Clip+)
14:55:58sideralthat's what we were talking about earlier, in a way
14:56:36sideralI am using Rockbox USB on the Clip+, but I've run into FS #11877
14:57:28[Saint]Hmmm...if I plug USB while RaaA is running on my phone, it reboots?
14:58:44ThomasAHsideral: so it works most of the time?
15:09:18 Quit soap (Read error: Connection reset by peer)
15:22:31 Join MatthewWilkes [0] (
15:22:40 Join Leif [0] (
15:24:29MatthewWilkesHey all, my Samsa Clip+ has decided to brick itsself (honest, it was working on monday, haven't used it since). When i turn it on I can get into rockbox 3.7.1 (no usb patch :() but when i try to boot original firmware it either goes to USB, gets to connected and then doesn't mount mass storage, or tries to refresh media, gets to 20% and reboots
15:24:45MatthewWilkesAny ideas of what I could do to gain access to the usb filesystem short of disassembling it?
15:26:15 Join soap [0] (
15:26:15 Quit soap (Changing host)
15:26:15 Join soap [0] (~soap@rockbox/staff/soap)
15:27:07gevaertsMatthewWilkes: I'm not a clip+ specialist so there may be better solutions, but you could always make a build with the usb patch and run that from an sd card
15:27:26MatthewWilkesgevaerts: how can I run from an SD card?
15:27:33MatthewWilkesDoes it 'just work'?
15:27:38gevaerts"play" the firmware file
15:28:13gevaertsYou won't get (well, not easily) working plugins and things like that that way, but you don't need that for USB
15:28:50MatthewWilkesHmm… I just put a microsd in and tried to go to files and it seems to have locked up
15:29:01MatthewWilkesactivity light is on though, is it doing something? :)
15:29:52*gevaerts thinks he'd better leave this to our AMS specialists...
15:30:13[Saint]That's not so good. RoLo/uSD is the only way I could think to get rockbox with USB on there.
15:30:43MatthewWilkesI have a 4gb and a 16gb µSD card
15:30:50MatthewWilkesboth SDHC, but the clip+ claims to support that
15:30:57gevaerts[Saint]: well, getting the OF not to be silly would also help presumably
15:30:57MatthewWilkesLocks up on boot with it in
15:31:06MatthewWilkesgevaerts: I called up sansa
15:31:12MatthewWilkesThey said post it to us and we'll give you a new one
15:31:18MatthewWilkesYay commercial software
15:31:22 Quit mystica555 (Ping timeout: 252 seconds)
15:32:20gevaertsDid you ever have either of those µSD cards working in it?
15:32:33MatthewWilkesnever tried them before
15:32:47MatthewWilkesphone's card and an old one, didn't think of it
15:32:58gevaertsright, so you might just be unlucky enough to have two cards that 3.7.1 doesn't like
15:32:58MatthewWilkesI have it set to refresh the database automatically, could it be that?
15:33:27gevaertsDisabling that can't hurt I guess
15:33:51MatthewWilkesoh, right, yeah, I can disable that
15:33:56 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
15:33:58MatthewWilkesrockbox didn't die :)
15:35:28 Join pamaury [0] (81680b01@rockbox/developer/pamaury)
15:37:23pamaurysideral: I think that you are violating the USB spec by using exclusively an OTG as host :)
15:38:06MatthewWilkesWiped the settings and i got a kernel panic when I inserted the sd
15:38:09MatthewWilkesnext time it mounted
15:38:18pamauryI must admit that I don't know much about it since we never tried to use an OTG device as host
15:40:14pamauryI would love to try but I have more important projects ;)
15:44:04gevaertsMatthewWilkes: right. I'd try building a rockbox.sansa from trunk and loading that from sd then
15:49:26MatthewWilkesgot usb build running, but the lappy isn't seeing it
15:49:27sideralThomasAH: Yes
15:49:59sideralpamaury: Same here
15:50:02ThomasAHsideral: good to hear
15:50:45amiconnpamaury: Isn't the only difference between plain host and otg that if you connect two otg devices they will negotiate which one will be host?
15:51:44pamaurythe difference is that by means of standard cables, you are only allowed to connect: host with OTG (used as device) or OTG with OTG (negociate)
15:51:50pamaurygevaerts: correct me if I'm saying something wrong :)
15:52:25gevaertspamaury: I'm not too familiar with OTG
15:52:38pamauryperhaps wikipedia helps, let's see....
15:52:44gevaertsMatthewWilkes: I'd say pamaury and sideral are our main specialists right now :)
15:52:57 Join DerPapst [0] (
15:53:47pamauryhum, I might be wrong
15:53:58pamauryI need to read the spec to be clear on that point
15:54:18*sideral is the specialist for contentious feature additions
15:54:24pamauryOne problem comes from the power delivered on the bus too
15:54:44MatthewWilkessideral: :D
15:54:49gevaertsMatthewWilkes: hm, have you verified that the cable is good?
15:55:37[Saint]jhMikeS: warnings while building with the new playback engine
15:55:39[Saint]/home/saint-ubuntu/Sources/rockbox/apps/playback.c: In function 'audio_init_codec':
15:55:39[Saint]/home/saint-ubuntu/Sources/rockbox/apps/playback.c:1161: warning: unused parameter 'track_info'
15:55:40MatthewWilkesgevaerts: I've used it for programming a microcontroller before, seems to work
15:56:02 Quit benedikt93 (Quit: The higher the IQ of the individual...The more they enjoy watching a robot drive into a wall at top speed)
15:56:49 Join pamaury_ [0] (81680b01@rockbox/developer/pamaury)
15:57:36 Join evilnick_B_ [0] (0c140464@rockbox/staff/evilnick)
15:58:28ThomasAHpamaury_: powered USB hub to the rescue :)
15:58:29jhMikeS[Saint]: erm...that sounds strange
15:58:38 Quit MatthewWilkes (Remote host closed the connection)
15:58:48 Quit pamaury (Ping timeout: 252 seconds)
15:58:48 Quit evilnick_B (Ping timeout: 252 seconds)
15:58:51pamaury_but then you have the problem of standard cables
15:59:03jhMikeS[Saint]: ah, ignore it. it's because codec buffering isn't supported
15:59:43pamaury_I've never see a micro(or whatever OTG compatible) / female A, you would need something like that
16:00:03jhMikeSit's supported, but only on native builds
16:00:31*pamaury_ decide to not talk about OTG until he knows what he is talking about :)
16:00:42 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
16:02:21 Nick evilnick_B_ is now known as evilnick_B (0c140464@rockbox/staff/evilnick)
16:02:23 Part thomasjfox
16:02:46 Join Rob2223 [0] (
16:04:41 Part mem_
16:05:33 Join MatthewWilkes [0] (
16:05:51MatthewWilkesOkay, I have a workaround that will satisfy the requirement of 'short haul flight in 2 hours'
16:05:57 Part LinusN
16:06:16MatthewWilkesBoot to rockbox without SD card in, insert SD, listen
16:06:42 Quit Rob2222 (Ping timeout: 276 seconds)
16:15:20 Quit liar (Quit: hallowed are the ori!)
16:16:32ThomasAHpamaury_: I have such a cable: male micro B to female A. Yes, it's non-standard
16:17:55ThomasAHpamaury_: I bought this one:
16:18:19pamaury_If I'm bored I could have atry. I have a TI-89 with a micro/micro cable and it's self-powered, perfect for trying otg with clip+ :) although I don't know which controller the clip+ has
16:18:51ThomasAHpamaury_: or here on instead of .de:
16:19:35 Join liar [0] (
16:19:36MatthewWilkesThomasAH: Indeed, I was just wondering why ebay was so clever to redirect me to the local site.
16:19:39ThomasAHpamaury_: ah, but I just notice that you're from Europe, too
16:19:52ThomasAHMatthewWilkes: *g*
16:20:11*MatthewWilkes germans it up a bit
16:20:14 Nick MatthewWilkes is now known as MatzeWilkes (
16:21:14ThomasAHbut of course Clip+ does not have a Micro B, but Mini connector
16:22:12MatzeWilkesyeah, and thankfully I had a mini connector for programming my
16:22:38 Quit evilnick_B (Ping timeout: 252 seconds)
16:22:38 Quit pamaury_ (Ping timeout: 252 seconds)
16:29:06 Quit Keripo (Ping timeout: 240 seconds)
16:30:46 Join ChickeNES [0] (~ChickeNES@
16:34:55 Join L-Strife89 [0] (~Strife89@
16:36:01 Quit skapazzo (Ping timeout: 264 seconds)
16:36:29 Join skapazzo [0] (~skapazzo@
16:39:34 Quit pepsi (Ping timeout: 252 seconds)
16:41:53 Join pepsi [0] (~pepsi@unaffiliated/jbutera)
16:46:23 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
16:48:03***Saving seen data "./dancer.seen"
16:50:28 Join archivator [0] (
16:50:31 Part archivator
16:50:35 Quit MatzeWilkes (Quit: This computer has gone to sleep)
16:52:21 Join Keripo [0] (
16:55:36[Saint]SVN head RaaA with list dpacing, meier crossfeed, new playback engine, resources-on-sd, audio dropout fix, quickscreen timeout, shutdown in main menu and "fancy" logos:
16:56:07[Saint]240x320, 320x480 and 480x800 builds in the same zip.
16:58:11 Quit sinthetek (Ping timeout: 240 seconds)
17:05:44 Quit Leif (Quit: Leaving)
17:09:52 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
17:32:33 Join stoffel [0] (
17:34:02 Join ender` [0] (
17:37:59bluebroth3rwhat are "fancy" logos?
17:38:05bluebroth3ras in bigger icons?
17:41:30[Saint]Nah, I gave the rockbox logo a bit of an update.
17:41:51[Saint]I made the "open source jukebox firmware" actually readable.
17:41:57*bluebroth3r wonders how that would look
17:42:34[Saint]Wonder no more ;)
17:43:39*jhMikeS thinks it's weird it's still just for the jukebox
17:44:19 Join komputes [0] (~komputes@ubuntu/member/komputes)
17:44:31*[Saint] assumed that "jukebox" took on a wider meaning ;)
17:45:48*jhMikeS gets some idea to emulate an archos jukebox in a plugin and run the original rockbox on that
17:47:35 Quit jfc (Ping timeout: 260 seconds)
17:48:49[Saint]pffffft! but instead you waste your time fixing up useless things like buffering and playback.
17:49:06 Quit kugel (Ping timeout: 240 seconds)
17:49:10[Saint]you should be ashamed jhMikeS, ashamed...
17:49:27 Quit pepsi (Ping timeout: 252 seconds)
17:50:48jhMikeSI should waste more time
17:51:04bluebroth3rtrue, who needs buffering or playback at all?
17:51:22bluebroth3rbtw, I find that "open source jukebox firmware" pretty readable on my phone :)
17:51:31jhMikeSnoone...totally useless stuff
17:51:47 Join kugel [0] (~kugel@rockbox/developer/kugel)
17:52:00 Join Buschel [0] (~chatzilla@
17:52:05 Join pepsi [0] (~pepsi@unaffiliated/jbutera)
17:52:53bluebroth3r[Saint]: btw, I would like to get a larger iconset into svn. You're still working on FS #11717?
17:53:28[Saint]bluebroth3r: I've not had the time to recently, but I'm hoping to get back into it shortly.
17:53:44[Saint]I'd like to get a larger iconset(s) into svn also.
17:53:58bluebroth3rok. Maybe I'll just start off with a generated one and let you post an update of that later on then :)
17:54:34[Saint]I have one in a task on flyspray somewhere...
17:55:21[Saint]It's a lot more honed that the generated ones, and it'll give me more of an excuse to fix it up ;)
17:55:50[Saint]In general, though, I'd like the underlying system of icons/viewers fixed before I work on a "final" iconset.
17:56:01[Saint]it's really quite busted presently.
17:56:19Buschelsmk: Just in case you read the logs -> I am interested in the changes you've made to libfaad (exchanging malloc with static vars). Is there any patch available?
17:58:01 Quit kugel (Ping timeout: 252 seconds)
17:58:16[Saint]bluebroth3r: Oh, is FS #11717 I was talking about.
17:58:59[Saint]Yes, I'd like to see that go in to help those that need a larger iconset. But I really *HATE* the way I worked around the fact that the underlying icon system is broken.
17:59:41[Saint]A big cleanup of viewers/icon code would be nice, but it's out of my league personally.
18:01:00 Quit ChickeNES (Quit: Computer has gone to sleep.)
18:02:18[Saint]What really needs to be done is to go through the iconset and establish what's being used, what's being used inconsistently, and how we *want* the icons (final iconset to be discussed) to be distributed.
18:02:26[Saint]ie. what the icons "mean".
18:02:39 Quit pepsi (Ping timeout: 252 seconds)
18:03:20 Quit Buschel (Ping timeout: 260 seconds)
18:03:38 Join madalu [0] (~user@unaffiliated/madalu)
18:04:18[Saint]a good 33% of that iconset is not used at all, and a lot of it is used rather inconsistently. This isnlt the fault of the iconset, I actually changed a lot of the icons to fit how they were being used...and, while that's not the fault of the iconset, it's very broken to work around it in this manner IMO
18:05:38 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
18:06:20[Saint]at some point in time people started going "oh, that icon looks nice...I'll use that", instead of using the icons for their intended purpose (apparently each icon had a purpose, I have a diagram that matches up the icon with the use case's rather horribly disregarded)
18:07:41[Saint]</rant> Sorry ;)
18:08:49jhMikeSI _think_ they had an intended purpose...some time in, what, 2006-2007 or so.
18:10:32[Saint]jhMikeS: Evidence I have suggests so ;)
18:10:53[Saint]It would be nice to go back to using them in a uniform way...that's why I *hate* my iconset.
18:11:05saratogam = (1<<16) / pow(2, (x+1) * (ENV_STEP/4.0) / 8.0);
18:11:09saratogawhere m is a double
18:11:16[Saint]It makes icons usable for large screen targets, but it's a fix for a broken system.
18:11:22saratogai can think of some reasons that VGM codec thing is slow . . .
18:11:53saratogaah that doesn't get compiled
18:11:58jhMikeSsaratoga: I did see float stuff in there
18:14:12saratoga double sample_rate
18:15:44 Quit soap (Read error: Operation timed out)
18:15:49 Join Horscht [0] (~Horscht@xbmc/user/horscht)
18:16:07 Join pepsi [0] (~pepsi@unaffiliated/jbutera)
18:17:15saratogai guess its a little premature to complain about that, since some of this code will probably change when the non-GPL stuff is dropped
18:17:25 Join soap [0] (~soap@rockbox/staff/soap)
18:18:36 Join saratoga_ [0] (9803c6dd@gateway/web/freenode/ip.
18:21:47 Quit saratoga (Ping timeout: 252 seconds)
18:22:31 Quit evilnick_B (Ping timeout: 252 seconds)
18:23:14 Quit bieber_ (Remote host closed the connection)
18:24:11jhMikeSguess we're a month in from its debut...wonder when the conversion will happen
18:24:50 Join Buschel [0] (
18:30:41saratoga_the NSF codec on the tracker also has a some nasty bits of non-GPL code built into it
18:31:29 Quit krazykit (Quit: awe yeeeeeee)
18:31:38 Quit pepsi (Ping timeout: 276 seconds)
18:32:09 Quit madalu (Remote host closed the connection)
18:34:21 Quit sideral (Quit: Leaving.)
18:34:59 Quit saratoga_ (Ping timeout: 252 seconds)
18:35:09 Join pepsi [0] (
18:35:10 Quit pepsi (Changing host)
18:35:10 Join pepsi [0] (~pepsi@unaffiliated/jbutera)
18:35:50 Quit L-Strife89 (Quit: Leaving)
18:38:57 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
18:39:04saratogaand Atari SC68 codec is GPLv3 only
18:42:46jhMikeSgama doesn't check these things?
18:48:05***Saving seen data "./dancer.seen"
18:48:32 Quit jhMikeS (Ping timeout: 240 seconds)
18:49:20 Join jhMikeS [0] (
18:49:25 Quit jhMikeS (Changing host)
18:49:25 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
18:52:57 Join ChickeNES [0] (~ChickeNES@
18:59:15*amiconn is still in favour of going gplv3
19:01:17 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
19:02:55 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
19:03:37 Quit DerPapst (Quit: Leaving.)
19:03:51 Quit kisak (Quit: Lost terminal)
19:03:52 Quit GeekShadow (Quit: The cake is a lie !)
19:12:27 Join kugel [0] (
19:12:27 Quit kugel (Changing host)
19:12:27 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:13:19 Quit stoffel (Ping timeout: 250 seconds)
19:14:29 Join stoffel [0] (
19:15:35 Quit Rob2223 (Remote host closed the connection)
19:15:51 Join Rob2222 [0] (~Miranda@
19:18:27 Quit [Saint] (Remote host closed the connection)
19:18:40 Quit kugel (Quit: leaving)
19:18:57 Join kugel [0] (
19:18:57 Quit kugel (Changing host)
19:18:57 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:20:22 Join [Saint] [0] (~st.lasciv@
19:21:21 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:24:05 Quit Keripo (Quit: Leaving.)
19:26:42 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
19:27:23 Join robin0800 [0] (
19:28:09 Quit [Saint] (Disconnected by services)
19:28:11 Join S_a_i_n_t [0] (~st.lasciv@
19:29:11 Join [Saint] [0] (~st.lasciv@
19:29:25 Quit S_a_i_n_t (Disconnected by services)
19:30:51 Quit stoffel (Remote host closed the connection)
19:33:49 Quit [Saint] (Disconnected by services)
19:33:51 Join S_a_i_n_t [0] (~st.lasciv@
19:36:38 Quit S_a_i_n_t (Client Quit)
19:36:57 Join Keripo [0] (
19:39:13 Join [Saint] [0] (~st.lasciv@
19:43:39 Quit logbot (Ping timeout: 250 seconds)
19:43:39***ERROR: (Closing Link: (Ping timeout: 250 seconds)) from
19:43:39***Saving seen data "./dancer.seen"
19:43:41***Started Dancer V4.16
19:43:41***Connected to on port 6667
19:43:41***Logfile for #rockbox started
19:43:41Mode"logbot :+i" by logbot
19:43:45***Server message 501: 'logbot :Unknown MODE flag'
19:43:45 Join logbot [0] (
19:43:45 Join [Saint] [0] (~st.lasciv@
19:43:45 Join Keripo [0] (
19:43:45 Join robin0800 [0] (
19:43:45 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
19:43:45 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:43:45 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:43:45 Join Rob2222 [0] (~Miranda@
19:43:45 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
19:43:45 Join ChickeNES [0] (~ChickeNES@
19:43:45 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
19:43:45 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
19:43:45 Join pepsi [0] (~pepsi@unaffiliated/jbutera)
19:43:45 Join Buschel [0] (
19:43:45 Join soap [0] (~soap@rockbox/staff/soap)
19:43:45 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:43:45 Join komputes [0] (~komputes@ubuntu/member/komputes)
19:43:45 Join ender` [0] (
19:43:45 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
19:43:45 Join skapazzo [0] (~skapazzo@
19:43:45 Join liar [0] (
19:43:45 Join dfkt [0] (dfkt@unaffiliated/dfkt)
19:43:45 Join TheLemonMan [0] (
19:43:45 Join leavittx [0] (~leavittx@
19:43:45 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
19:43:45 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
19:43:45 Join n1s [0] (~n1s@rockbox/developer/n1s)
19:43:45 Join pixelma [0] (quassel@rockbox/staff/pixelma)
19:43:45 Join amiconn [0] (quassel@rockbox/developer/amiconn)
19:43:45 Join Strife89 [0] (~Strife89@
19:43:45 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
19:43:45 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
19:43:45 Join simonlnu [0] (simon@unaffiliated/simonrvn)
19:43:45 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
19:43:45 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
19:43:45 Join avacore [0] (
19:43:45 Join bluefoxx [0] (
19:43:45 Join balintx [0] (
19:43:45 Join mystica555_ [0] (
19:43:45 Join amee2k [0] (
19:43:45 Join eGen_ [0] (
19:43:45 Join sasquatch [0] (~username@
19:43:45 Join Elfish [0] (amba@2a01:4f8:100:90a1:abc:abc:abc:abc)
19:43:45 Join mudd1 [0] (
19:43:45 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
19:43:45 Join Zarggg [0] (
19:43:45 Join slooopy [0] (
19:43:45 Join markun [0] (~markun@rockbox/developer/markun)
19:43:45 Join gromit` [0] (
19:43:45 Join B4gder [0] (~daniel@rockbox/developer/bagder)
19:43:45 Join timccc [0] (~timccc@
19:43:45 Join Buganini [0] (~buganini@2001:c08:2600::379)
19:43:45 Join YPSY [0] (
19:43:45 Join yosafbridge [0] (
19:43:45 Join mikroflo1s_ [0] (
19:43:45 Join Torne [0] (~torne@rockbox/developer/Torne)
19:43:45 Join FoolOnHill [0] (
19:43:45 Join mt [0] (~mtee@rockbox/developer/mt)
19:43:45 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
19:43:45 Join [fred] [0] (
19:43:45 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
19:43:45 Join zu_ [0] (~zu@
19:43:45 Join jae [0] (
19:43:45 Join n17ikh [0] (
19:43:45 Join Barahir [0] (
19:43:45 Join alexbobp [0] (
19:43:45 Join Galois [0] (
19:43:45 Join fyrestorm [0] (
19:43:45 Join sobel [0] (~matt@
19:43:45 Join z35 [0] (~z35@
19:43:45 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
19:43:45 Join froggyman [0] (~seth@unaffiliated/froggyman)
19:43:45 Join Zambezi [0] (Zulu@
19:43:45 Join JesusChrysler [0] (
19:43:45 Join cjcopi [0] (
19:43:45 Join scorche [0] (~scorche@rockbox/administrator/scorche)
19:43:45 Join AlexP [0] (~alex@rockbox/staff/AlexP)
19:43:45 Join niekie [0] (~niek@CAcert/Assurer/niekie)
19:43:45 Join bieber [0] (
19:43:45 Join tmzt [0] (~tmzt@
19:43:45 Join rasher [0] (~rasher@rockbox/developer/rasher)
19:43:45 Join Guinness [0] (
19:43:45 Join maraz [0] (
19:43:45 Join swilde [0] (
19:43:45 Join martii [0] (
19:43:45 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
19:43:45 Join pikytcus [0] (
19:43:45 Join Xerion [0] (
19:43:45 Join logvelc [0] (
19:43:45 Join scorche|sh [0] (~scorche@rockbox/administrator/scorche)
19:43:45 Join dionoea [0] (~dionoea@videolan/developer/dionoea)
19:43:45 Join linuxguy3 [0] (
19:43:45 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
19:43:45 Join Rondom [0] (
19:43:45 Join simabeis_ [0] (
19:43:45 Join jepler [0] (~jepler@emc/developer/pdpc.professional.jepler)
19:43:45 Join kkit|sh [0] (
19:43:45 Join bthomson [0] (
19:43:45 Join pjm0616 [0] (~user@
19:43:45 Join feisar-_ [0] (
19:43:45 Join aevin [0] (eivindsy@unaffiliated/aevin)
19:43:45 Join knittl [0] (~knittl@unaffiliated/knittl)
19:43:45 Join guymann [0] (
19:43:45 Join Battousai [0] (~bryan@gentoo/developer/battousai)
19:43:45 Join tguinot [0] (
19:43:45 Join bzed [0] (~bzed@2001:6f8:118a::100)
19:43:45 Join ack` [0] (
19:43:45 Join elcan [0] (
19:43:45 Join preglow [0] (thomj@rockbox/developer/preglow)
19:43:45 Join crwl [0] (
19:43:45 Join parafin [0] (
19:43:45 Join TBFOOL [0] (
19:43:45 Join literal [0] (
19:43:45 Join CIA-87 [0] (~CIA@
19:43:45 Join Farthen [0] (
19:43:45 Join merbanan [0] (
19:43:45 Join ved [0] (
19:43:45 Join Utchy [0] (
19:43:45 Join FOAD [0] (~dok@
19:43:45 Join jvoisin [0] (
19:43:45 Join Hadaka [0] (
19:43:45 Join ThomasAH [0] (
19:43:45 Join olejorgenb [0] (
19:43:45 Join @ChanServ [0] (ChanServ@services.)
19:43:45 Join ranmachan [0] (
19:49:33bluebroth3r[Saint]: see FS #11982 for what each icon means −− at least according to the enum in the code :)
19:50:01bluebroth3rbut anyway, I really think a bigger icon set is needed. The tiny ones look really displaced on a large / high resolution screen
19:50:29bluebroth3rso regardless of the systems brokenness I'd like to get that in, in one way or another :)
19:51:16[Saint]Yes, I'm not opposed to it going in. I'm opposed to to it being a "fix" for anything though ;)
19:51:26[Saint]Just a stop-gap measure.
19:52:21bluebroth3rI'm planning a stop-gap fix for the TTS issue in Rockbox Utility on multicore Windows machines as well. Just so we can make a new release −− the unzip issue really needs it
19:52:48[Saint]Yeah, I saw another issue with that in the forums.
19:53:08[Saint]I wonder how many themes are using "broken" zip formats
19:54:01bluebroth3ralso, I was thinking that it would be good to put at least the original sources for various icons in a common place. Like rockbox-clef.svg which is used (converted to png) as icon on Android, and directly in Rockbox Utility.
19:54:09 Join wodz [0] (
19:54:18[Saint]I've *always* used, people have probably never been able to use my themes with RbUtil. Seems kinda funny it took so long for someone to notice.
19:54:44[Saint]bluebroth3r: Yep, I'd like that.
19:55:18bluebroth3rthen I've made the svg version of the "Rb" logo that isn't use anywhere right now. rockbox.ico (used in Rockbox Utility) uses it but the ico file is not based on the svg
19:55:27wodzhow to split data section into part in dram and part in iram in linker script? define additional section?
19:55:31bluebroth3rand we have the "full" logo as svg that is inside of the manual tree
19:56:42[Saint]Oh, regarding my "fancy" logo, I'm using that because it makes it readable on the "square" series of widgets in RaaA 240x320
19:56:56[Saint]also because I like playfully using the android logo font
19:57:06bluebroth3rwodz: I've never did that myself but I think that's the way to do it.
19:58:12bluebroth3r[Saint]: any idea where to place a common "original artwork" folder?
19:59:22[Saint]bluebroth3r: Hmmm...looking at the tree....not off the top of my head :/
19:59:39bluebroth3rthat might also be a good location to put the icon strip creation script at :)
20:00:34[Saint]perhaps it deserves it's own folder then...and maybe we can dig up some sources for cabbie and/or move them there also.
20:00:58[Saint]I _think_ there's *some* cabbie source images in svn
20:02:34[Saint]aha...yes, there's some .xcf (ew) files in wps\cabbiev2\sources
20:03:25bluebroth3rwhat's the problem with xcf? ;-)
20:03:39bluebroth3roh, and s/\\/s/g :)
20:03:58bluebroth3rargh, s/\\/\//g
20:10:46pixelmakugel, jhMikeS: I'm using a playback rework patched build on my phone (simply listening during commute) and one thing I encountered but *believe* to have seen it before is that if I let the playlist play to the end and then chose "Resume" from the menu, it doesn't splash "nothing to resume" but instead plays the last track from the start. Maybe kugel would have an explanation if it also happens in SVN or not (in case buffering on Android does
20:10:46pixelmasomething differently) but I'll also try without the patch and on other targets as well later
20:12:03[Saint]pixelma: I posted an updated (sync'd playback engine build) .apk a while back in the logs
20:12:12[Saint](if you need it, that is)
20:13:27pixelmaI once had Rockbox loading hang with your apk (while it had to unzip itself or whatever it does, showing the xx/yy), not sure yet if it's reproducible or not
20:14:18[Saint]Hmmm, I've not seen that myself. I've only seen that with the "non-resources-on-uSD" builds.
20:14:36[Saint]Sometimes the first load of the application results in just a black idea why.
20:20:05 Quit ChickeNES (Quit: Computer has gone to sleep.)
20:20:36jhMikeSpixelma: haven't encountered that. It nullfies resumes when it stops but maybe that fails to get saved there.
20:21:19pixelmaas I said, I'm not sure if it's related to your patch or not... will try after dinner
20:25:32jhMikeSi've tested end of playlist on the targets I have and it ends it properly (just tried it now too)
20:26:23[Saint]give the poor girl a chance to test a more recent build, geeez! :P
20:27:10jhMikeSforget dinner, test patches!
20:30:20 Join ChickeNES [0] (~ChickeNES@
20:31:38jhMikeSone thing's for sure, the mp3 encoder codec hates 320kpbs @ 32KHz
20:32:32 Quit GodEater (Remote host closed the connection)
20:33:54bluebroth3rpixelma: any thoughts on switching the manual to Latin Modern fonts? They look pretty much the same as the currently used Computer Modern but work better when searching in the pdf (and they "fix" this "bug" FS #12057)
20:34:24 Quit leavittx (Ping timeout: 260 seconds)
20:34:29[Saint]heh..."bug", yeah :/
20:35:21jhMikeSdoesn't like 256@32KHz either, < 224 @32k seems to work
20:35:44bluebroth3rthat reporter clearly confuses vector fonts with anti aliased fonts
20:36:06[Saint]I was actually thinking that.
20:36:17[Saint]Shit, it's a can read it.
20:36:40[Saint]I would have been tempted to close that out as "notabug" myself...
20:36:51bluebroth3rwell, he has a point: vector fonts can get zoomed more nicely. But having a bitmap font doesn't mean you can't anti alias it
20:37:36bluebroth3rand calling bitmap fonts as "only exist because of the early history of computing" is ... well. Seems Rockbox itself is doing it wrong then.
20:37:41[Saint]why would you *want* to zoom it to a point where it looks obviouslt terrible anyway?
20:37:49bluebroth3rplus only exist because of the early history of computing
20:38:10[Saint]if you're that blind, get glasses, if you have glasses and still can't see it properly, use a screen reader ;)
20:38:24bluebroth3rbesides, it's completely pointless to zoom in the manual to have a usual screen only fit like 5 characters. That's ... useless
20:38:26jhMikeScomputers exist only because of the early history of computing
20:38:33bluebroth3rjhMikeS: indeed!
20:38:55[Saint]bluebroth3r: yes, I think we both just said the same thing in different ways.
20:39:03[Saint]why *would* you zoom in that far?
20:39:03jhMikeSwhich I think began with sheep counting :\
20:39:07bluebroth3rI was tempted to close that as WONTFIX a couple of times :)
20:40:01[Saint]"I zoomed in really far in the manual and I saw some square edges, it offended my deleicat sensibilities...this is a bug"
20:40:08bluebroth3rI gave it a look anyway (when looking for LaTeX / TeXLive on Cygwin). Turns out that lmodern has advantages in searching. I don't see advantages because it's "more anti aliased"
20:40:22 Quit TheLemonMan (Read error: No route to host)
20:41:03[Saint]Close his task as WONTFIX, then open your own and submit that ;)
20:41:26[Saint]it's clearly different issues ;)
20:42:45 Join TheLemonMan [0] (
20:43:47 Quit merbanan (Ping timeout: 276 seconds)
20:43:51bluebroth3rI was in fact considering something like that.
20:44:39bluebroth3r(if only because I'm pissed by those arguments. I asked why bitmap fonts are a problem and he lists general advantages of vector fonts. Argh)
20:44:54bluebroth3rthough doing so might be a bit immature :)
20:45:04bertriksounds a bit like the git vs. svn discussion :)
20:45:43[Saint]I read the task, and, I'm still not certain he actually answered (or in fact understood) your question
20:47:07bluebroth3rbertrik: you've got a point, indeed. At least to some degree :)
20:48:33saratogawe need to switch to a new version control system, one more complicated
20:49:15bluebroth3rit has to be obscure too!
20:50:04saratogalets just use SQL for version control
20:50:27bluebroth3rhow about using Excel?
20:50:48bluebroth3rand use Word for programming
20:50:53 Join webguest [0] (
20:52:51 Quit webguest (Client Quit)
20:53:19 Quit Keripo (Quit: Leaving.)
20:58:25 Join merbanan [0] (
21:04:09tguinotLotus notes is my favorite IDE
21:06:20linuxstbbluebroth3r: Something must have changed with the manual, as the 3.0 manuals for example have much better quality typography. They used vector fonts, but also a different font, and the typography is better in the 3.0 manual.
21:07:42 Join Keripo [0] (
21:07:56linuxstbbluebroth3r: Also, regarding bitmap fonts, it's not just a matter of zooming in, it's the artifacts you can see when you do any kind of resizing.
21:08:08*linuxstb admits to using evince though - maybe acroread does a nicer job
21:10:42 Join mshathlonxp [0] (
21:15:05 Quit mudd1 (Ping timeout: 240 seconds)
21:15:33pixelmabluebroth3r: if it works (also building the manual) and doesn't bring back the dreaded ligature bug again, I don't see the reason not to
21:15:53pixelmaa reason
21:19:31pixelmajhMikeS: you enabled test_codec in your v5 patch (unconditionally too) - just wondered why a OndioFM build would fail on test_codec ;)
21:19:36 Join L-Strife89 [0] (~Strife89@
21:22:41 Join ChickeNE_ [0] (~ChickeNES@
21:22:44 Quit ChickeNES (Read error: Connection reset by peer)
21:23:19jhMikeSpixelma: I'd think it should, yes :)
21:24:44jhMikeStest_codec was enabled in all the patches, actually
21:26:42pixelmameh, no Android testing for me today anymore - I forgot my microUSB cable at work and the battery is already quite flat
21:27:18jhMikeSalways check the battery pressure regularly :D
21:29:25pixelma7% :P
21:29:27bluebroth3rthose smartphones are eating up battery like mad
21:30:04bluebroth3rpixelma: well, I can search for ligatures (like ffi) just fine. That's with evince on Linux, maybe I should try the pdf with Adobe Reader
21:30:30[Saint]govern the CPU properly with setCPU profiles, and set up tasker to do things sanely...and it can actually last a damn while.
21:30:56[Saint]I can get a full day out of my phone easily now with heavy use.
21:31:09pixelmait's a bit offtopic-y here though, I just mentioned it because I wanted to do some patch testing
21:32:08bluebroth3rlinuxstb: I don't see artifacts during resize with Evince
21:32:12 Quit Buganini (Remote host closed the connection)
21:35:16 Join Buganini [0] (~buganini@2001:c08:2600::379)
21:35:48linuxstbbluebroth3r: At 200% for example, I see two noticable things - horizontal lines in letters (e.g. in a lower-case e) look fainter than the rest of the letters, and the bottoms of the letters don't always line up properly.
21:36:49 Quit skapazzo (Ping timeout: 260 seconds)
21:38:07 Join GodEater [0] (
21:38:07 Quit GodEater (Changing host)
21:38:07 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
21:40:42bluebroth3rlinuxstb: you're right. I was using evince as well, but with magnification of 400% rather than 200% :)
21:41:11bluebroth3rand the lmodern variant looks like crap in Adobe Reader if I disable "Smooth Text" (that's Reader 8.0)
21:41:28bluebroth3rthe cm variant seems to be rather strangely rendered
21:42:10bluebroth3r(without "Smooth Text" that is)
21:42:27bluebroth3rso the rendering is at least an important factor on this
21:43:45***Saving seen data "./dancer.seen"
21:48:47pixelmaI also think that linuxstb is correct in thinking the font changed at some point (and I also liked the old one a bit better overall), but I don't consider this very important if it is still readable
21:49:24bluebroth3rwe changed the font from palatino to computer modern because of small caps / ligature issues (can't remember which exactly it was)
21:49:33 Quit ChickeNE_ (Quit: Computer has gone to sleep.)
21:50:47bluebroth3rzooming into the "old" manual looks pretty good with SumatraPDF. It looks worse with Adobe Reader 8.0
21:51:26pixelmaah yeah, and now the font in some the player graphics doesn't fit anymore (I remember specifically chosing a Palatino family one for the ones I made)
21:51:31bluebroth3radditionally, I can find Bjoern in the credits (with umlaut) with SumatraPDF while Adobe Reader doesn't find it.
21:52:09 Join skapazzo [0] (~skapazzo@
21:52:34bluebroth3rwith lmodern it works with Adobe Reader too
21:52:52pixelmasounds good then
21:53:09bluebroth3rbut from what I can see this looks like Adobe Reader is rendering (and searching) crap here.
21:53:42 Quit L-Strife89 (Ping timeout: 240 seconds)
21:53:44pixelma8.0 is quite old though, have you tried a newer version?
21:54:39bluebroth3rI can try using the newer replacement for palatino.sty and see if that has the same advantages. If yes we can switch to palatino again −− I don't care too much about the font (though I have to admit that I like Computer Modern :)
21:55:09bluebroth3rcan I install different versions in parallel? I was always annoyed by Adobe Reader getting bigger and bigger (which is the reason I stayed at 8.0)
21:55:41bluebroth3roh, it's 8.2.6 to be exact. I did install at least the updates to 8 :)
21:58:19pixelmawell, I don't know why - usually I prefer sans-serif fonts but somehow still liked the Palatino one a bit better. As I said, I don't care much either and I don't think it's necessary to switch back if it's more work
21:58:27 Join leavittx [0] (~leavittx@
21:58:51 Join ChickeNES [0] (~ChickeNES@
21:59:19pixelmabluebroth3r: if you provide a downloadable version, I could quickly try in AdobeReader X
22:01:43bluebroth3rpixelma: both variants at
22:06:20 Join mudd1 [0] (
22:07:18pixelmathe look is quite different when zoomed to "full page" - the CM one looks "thinner"
22:09:51*bluebroth3r tries the packages recommended as replacement for palatino
22:11:07pixelmaand the CM one has the ligature problem (e.g. I can find "efficient" in the lmodern but not in the CM version) as well as umlauts are there but also not searchable
22:11:49pixelmabut in the lmodern PDF I can also search for words with umlauts
22:12:31wodzTheSeven: ping
22:12:51bluebroth3rpixelma: mathpazo one (palatino replacement as recommended by l2tabu.pdf) uploaded.
22:13:36TheSevenwodz: gnip
22:13:57bluebroth3rI can search for umlauts, but small caps only work if I search for "G eneral S etting"
22:14:05wodzTheSeven: what is used to do system_reboot() on nano2g? watchdog?
22:14:19TheSevenno, some kind of reset controller
22:14:38TheSevenmight be in the same IP as the watchdog, but can be triggered directly
22:14:50TheSevenjust write that value to that reg and it will reset immediately
22:15:47wodzTheSeven: one more thing is INT_TIMER() used anywhere?
22:15:52TheSeventhe dead loop after it is just to prevent the cpu from prefetching crap
22:15:53*bluebroth3r tries using the soul package for PDF output as well
22:16:07TheSevenwodz: yes, in system-s5l8700.c
22:16:42wodzTheSeven: in system-s5l8700.c it is defined but I can't see how it is used
22:16:42TheSevenit isn't used on the 8700 (which has dedicated IRQs for the timers), but on the 8701/8702/8720 (which have one shared irq)
22:16:57 Quit Keripo (Quit: Leaving.)
22:16:58wodzI see
22:16:58TheSevenit's in some irq vector table further down
22:17:32wodzTheSeven: now I see
22:17:39 Join Keripo [0] (
22:18:03wodzis it better to use hardware to do reset or simply jump into reset vector?
22:18:23 Quit Keripo (Client Quit)
22:20:33pixelmabluebroth3r: looks to me like there are different fonts in the headline and the standard text? Maybe that's why I found the fonts odd
22:21:20pixelma(currently comparing the two vector font versions and I think it's in both)
22:22:09bluebroth3rTeX capacity exceeded. Ouch.
22:22:42pixelmabut switching to vector would be an overall improvement here it seems for the above mentioned reasons not only font style
22:22:48 Quit slooopy (Ping timeout: 246 seconds)
22:23:11bluebroth3ryes, the main reason for switching would be the improved searching.
22:23:53bluebroth3rI don't care much about the fonts being bitmap, but I didn't remember that searching for umlauts didn't work.
22:24:44bluebroth3rsince lmodern work for small caps and umlauts (and mathpazo doesn't for small caps) I'd go for that.
22:24:57 Quit ChickeNES (Quit: Computer has gone to sleep.)
22:25:23pixelmafine with me
22:26:06bluebroth3rif someone finds a solution to make that work with mathpazo I won't object either but I'm not too motivated to look that much deeper into it.
22:27:41pixelmasearchability is much more important and the rest doesn't really matter (and I can't tell which I like better now anymore)
22:28:57*bluebroth3r agrees
22:29:02AlexPDoes comic sans work?
22:29:11*bluebroth3r slaps AlexP
22:29:17*[Saint] slaps alexbobp
22:29:22[Saint]AlexP too ;)
22:29:26bluebroth3rwith ... a Comic Sans character?
22:29:30[Saint]sorry alexbobp
22:29:30AlexP[Saint]: Lashing out at random? :)
22:29:32bluebroth3ror maybe use all :)
22:29:58[Saint]AlexP: it's your fault for not being the first in the nick complete queue :P
22:30:21pixelmait is here :P
22:30:56AlexPbluebroth3r: I agree :)
22:32:18bluebroth3rhmm. mathpazo doesn't do small caps as it seems.
22:33:11 Join ChickeNES [0] (~ChickeNES@
22:33:43bluebroth3rwhy on earth does evince not show the filename of the file that's open?
22:34:37pixelma[Saint]: did you ever try jhMikeS' patch on your phone without the "fix audio stopping"?
22:35:36[Saint]pixelma: yes, and it still cuts out.
22:35:51pixelmaah, that's what I wanted to know
22:35:59*bluebroth3r isn't surprised by that
22:41:29[Saint]Sorry alexbobp ;) It's AlexP's fault, really.
22:51:05pixelmajhMikeS: fwiw, I can't reproduce the resume "problem" with my c250 (patched build and Rockbox... ermm... 3.7 in comparison). That was just a quick test though, I'd like to run with it for a while as I was getting "playback randomly skips to the next track" before with 3.7 but it's very unpredictable
22:55:21 Quit GodEater (Read error: Operation timed out)
22:56:06bluebroth3rhmm, searching the mathpazo pdf with Reader X seems to not have issues with small caps
22:59:49bluebroth3rsame with Reader 8, but not evince or xpdf
23:00:33bluebroth3rSumatraPDF does work again
23:00:50AlexPbluebroth3r: What's an example phrase to search for?
23:01:02bluebroth3r"General Settings"
23:01:58bluebroth3rcheck if it finds the small caps versions of it −− like in 5.7.3 (which is nice because it has the text both in the section title and directly below in the text itself)
23:02:22AlexPbluebroth3r: Whatever chrome is using to display pdfs in page is working, okular doesn't
23:03:45bluebroth3rthe funny thing is SumatraPDF: AFAIK it's based on poppler, but evince is based on poppler as well ...
23:04:04bluebroth3rmaybe it works with never versions. That could also explain why it's not working with xpdf
23:06:27bluebroth3rwith lmodern all works in xpdf, evince, SumatraPDF and Reader.
23:06:42bluebroth3rwhy on earth does this have to be that much of a pain?
23:11:04AlexPbluebroth3r: lmodern works in okular too
23:12:41 Quit mshathlonxp (Quit: Leaving)
23:13:29pixelma[Saint]: a bit unrelated - but do you use touch buttons for seeking in any of your themes and have you tried those recently?
23:13:45[Saint]Yes, and yes.
23:14:18[Saint]I think I might have asked you...but, do you have any desire to give my RaaA-Cabbie a go?
23:14:19pixelmafor me it seems to start and immediately stop seeking
23:14:58[Saint]Oh...Hmmm...there was a change to touch actions, hang on a sec.
23:15:22pixelmaI use two touch areas in the same spot - short for skipping and holding for seeking. It worked some time ago
23:15:35bluebroth3rspeaking about cabbie ... anyone thought of moving the fwd and back buttons to the right and left of the album art?
23:15:45[Saint]pixelma: actually...I see it too.
23:15:57bluebroth3rright now they are rather close together. I find that somewhat too close
23:15:59AlexPbluebroth3r: I stuck touchscreen/android cabbie on the devcon agenda
23:16:07AlexPWe need to have chat about what we want
23:16:09bluebroth3rAlexP: good idea :)
23:16:22bluebroth3rand maybe a whiteboard to sketch it :)
23:16:38AlexPIt is a more important default there, and I think we need to stick a bit more closely to accepted conventions
23:16:43pixelma[Saint]: not *much* interested but for the sake of testing in case that should become default. Take your time though
23:16:57[Saint]pixelma: it's FS #29653
23:17:03bluebroth3rconventions as in Android conventions?
23:17:23[Saint]it needs a * instead of a & if it's supposed to be a repeat instead of a single action
23:17:29[Saint]pixelma: ^
23:17:46AlexPbluebroth3r: Touchscreen in general
23:18:04pixelma[Saint]: oh, when did that change?
23:18:10[Saint]so, if your ffwd/rew is a hold action, it needs an * instead of a & now
23:18:16AlexPbluebroth3r: First example of this was showing what clicking on an icon will do, and not it's current state (play/pause)
23:18:27[Saint]pixelma: ^
23:19:06 Quit bertrik (Remote host closed the connection)
23:19:10 Quit ChickeNES (Quit: Computer has gone to sleep.)
23:19:17[Saint]I accidentally put an old theme zip on my phone, so I forgot about it too ;)
23:19:36[Saint]but yes, there was a change to allow a single action to fire on a long press.
23:19:55pixelma[Saint]: ok, thanks for the info. I'll try as soon as possible. I also remember that I couldn't get the labelled touch area timeouts to work shortly after they were introduced, got to try again
23:20:11[Saint]otherwise if you had next/prev or mute on a long press, it cycled through them at the button_repeat rate :P
23:20:15 Join bertrik [0] (
23:20:15 Quit bertrik (Changing host)
23:20:15 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
23:20:40[Saint]pixelma: it's not labled's timeouts in particular for touch areas.
23:20:44[Saint]they're busted.
23:20:44bluebroth3rAlexP: good point
23:20:54[Saint]they always seem to default to 2 seconds (or less?)
23:20:58pixelma[Saint]: huh, I wonder though... stop as a long press action still works for me
23:21:26pixelmawithout changing to *
23:21:33[Saint]pixelma: Hmmm...that's interesting. Same here. I guess it's just a feature of stop.
23:22:48pixelmawell, I guess the commit message explains it
23:23:18pixelmaas stop is something that only fires once usually... thanks for the pointer, I didn't notice
23:23:52bluebroth3rwhat LaTeX distribution and version is the build server using?
23:26:10 Join ChickeNES [0] (~ChickeNES@
23:28:36 Join DerPapst [0] (
23:31:01 Quit bertrik (Quit: Leaving)
23:31:19 Join bertrik [0] (
23:31:19 Quit bertrik (Changing host)
23:31:19 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
23:32:31[Saint]pixelma: Yes, sorry. I confused myself. Yes, that makes perfect sense.
23:33:11[Saint]The reason it was changed was to allow me to put mute on a long press (to lessen accidental presses) and not have it fire on/off at the button repeat rate.
23:33:21[Saint]which in RaaA seems to be very high
23:34:04[Saint]Without the change if mute was set to a long press it was a real gamble trying to un-/apply mute
23:34:10 Quit komputes (Remote host closed the connection)
23:39:13 Quit Stummi (Quit: Bye!)
23:43:47***Saving seen data "./dancer.seen"
23:43:52 Join fdinel [0] (
23:45:16 Quit evilnick_B (Quit: Page closed)
23:46:51 Quit robin0800 (Quit: Leaving)
23:48:51 Quit n1s (Quit: Lmnar)
23:50:43CIA-87New commit by bertrik (r29755): iap: split iap_handlepkt into a function per mode
23:52:31CIA-87New commit by bluebrother (r29756): Use Latin Modern font family for the manual. ...
23:54:43CIA-87r29755 build result: All green
23:58:41CIA-87r29756 build result: All green

Previous day | Next day