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 2011-10-17

00:00:18***Saving seen data "./dancer.seen"
00:07:00 Join Strife89 [0] (
00:22:59 Quit bertrik (Ping timeout: 240 seconds)
00:24:25JdGordonr30762 fixes nothing, suggesting it does is pretty poor... it a a stopgap solution
00:28:47KiwiCamYou know how you can set Skip To Outro. Is there a tag for this in the Theme specs? I can't find one, but I had an idea for its use. I can replicate it using the %pE, but it would be nice to be able to base the pE value on the already selected Outro Time.
00:36:49 Quit Strife89 (Ping timeout: 258 seconds)
00:40:30 Join Scromple [0] (
01:00:56JdGordonKiwiCam: you can use any setting in a tag
01:00:57KiwiCamAlso. Skin Buffer. Could someone please clarify what the purpose of the Skin Buffer is?
01:01:17JdGordonits either %St or %Sx, cant remmeber which
01:01:26JdGordonthe skin buffer is what the tags are loaded into
01:01:54JdGordon%pE wont accept another tag as the timeout though, but that can be chaned easily
01:03:17 Join Strife89 [0] (~Strife89@
01:04:19 Quit ender (Quit: Outside of a dog, a book is a man's best friend. Inside of a dog it's too dark to read. -- Groucho Marx)
01:05:02KiwiCamJdGordon: I'll have a look through the %S stuff again. I like the pE tag, but it'd be nice to make use of the Outro time. ie If the Outro time is set to 30 Seconds, then pop up an indicator which means "You can press next now, if you like".
01:06:25JdGordonit wont "just work", and wont even parse now if you try to use it
01:09:26KiwiCamCan you please point me to a list of the actual setting names?
01:10:10KiwiCam"it wont "just work", and wont even parse now if you try to use it" I didn't follow that sorry.
01:11:27JdGordonsetting names can be found if you use the setting dumper plugin
01:11:43JdGordonthat was saying, there is no point playing with it thoughu as it wont work
01:12:03JdGordon%pE(%St(outro)) just wont work
01:12:19KiwiCamAs for the Skin Buffer. Is it like a storage place for the values of the tags? How is the limit reached?
01:13:14JdGordonit is just generic storage for the skins... the more tags you use (and viewports, etc) the more you will use
01:13:22KiwiCamOK. I'll just use the pE for now.
01:15:33KiwiCamAh. That's what I was looking for :) Now I see that I can create a text file called skin_buffer_size.txt and put a value in it. Say 60 (Clip's default is 48kb I understand). Does that text file work? It doesn't seem to in the emulator.
01:16:25JdGordonno, that was just removed
01:16:58JdGordonso update to svn and if your theme still doesnt load there is nothing you can do
01:17:31KiwiCamI am disappoint :(
01:18:44KiwiCamHas the buffer size been increased for the Clip in the latest svn (which I've never used and don't know where to start)?
01:18:45JdGordonyou amn't the only one
01:19:07JdGordonyes, the commit on the frontpage mentions it
01:19:18JdGordonthere is a bit of an increase
01:20:40KiwiCamI'm addicted to theming but I'm bumping against the buffer all the time. It's amazing what the clip can produce, but I could grind out some real creations with just a tad more space.
01:21:11JdGordonwhat is it with kiwis and themeing? :)
01:21:23JdGordoni need to start working on my clip skin
01:21:31JdGordongive you enough to finish it :)
01:21:44JdGordondo you know how to code?
01:22:11KiwiCamI'll start looking into svn. It's just that the themeing takes enough of my time now, so getting into that would be hazardous to my health ;) (and sanity). Will the buffer increase be included in the next Utility update at the end of this month?
01:23:16 Join kadoban [0] (
01:23:36LloreanJust download the "current build" instead of the last release, and you'll have the buffer increase.
01:23:52KiwiCamJust themes. I've just uploaded my second.
01:24:03LloreanThe "current build" reflects the current state of Rockbox SVN.
01:27:37KiwiCamOh wow. As simple as that. Here's me thinking that I'd be needing to download source, make and all that. So, if I make a bass ass theme and upload it, I just need to state that it will only work on the latest svn?
01:28:40LloreanAssuming it actually *does* work on the latest SVN, yeah.
01:28:53LloreanYou may wish to look at what revision you tested it on, and state it works on that one or later.
01:29:16LloreanIdeally the theme site should check whether themes work or not, but I'm not sure if the current problem is one that it does (or can) test for. It may be able to, I don't know.
01:31:34KiwiCamThanks Guys! I'll leave you to it now.
01:34:22JdGordonKiwiCam: are you just doing wps themeing? you know you can theme the way lists are drawn too?
01:37:26 Join ChickeNES [0] (~ChickeNES@
01:38:31 Quit ChickeNES (Client Quit)
01:38:49 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
01:40:42KiwiCamYes. I'll delve into that soon. It's something haven't tried yet. I have a few ideas and one question, but I need to do some testing first before bothering you with that.
01:42:52 Quit Strife89 (Quit: Night.)
01:44:27KiwiCamI haven't actually ventured into the sbs etc, because of the buffer issue. I'm maxing out the wps and so for me it felt pointless to start making sbs's etc, that I couldn't really use.
01:52:46 Quit mudd1 (Ping timeout: 248 seconds)
01:53:57 Quit simonlnu (Quit: .)
02:00:20***Saving seen data "./dancer.seen"
02:03:39 Join philipballew [0] (~philipbal@ubuntu/member/philipballew)
02:04:19philipballewwhats the best tool to update my rockbox ipod with linux?
02:05:00JdGordonanything that can unzip a file
02:06:18philipballewalright. Im using rythembox, but it sems to be acting funny
02:06:27philipballewwell my ipod that is
02:06:44philipballewit has the files on there, but rockbox wont show it
02:07:47JdGordonrockbox wont show "it" what?
02:08:28philipballewrockbox does not show my library of music on my ipod
02:08:50JdGordonwhere have you got the music stored?
02:08:51KiwiCamTransparency. I recall a forum thread, or a comment somewhere about transparency for the Clip being a little known secret. Is it true you can use it with the Clip? If so, what would be a practical example of its use?
02:09:00JdGordonare you using the ipod database and files?
02:09:10JdGordoni tihnk it makes the oflder hidden by default
02:10:33philipballewim just updating the default way rythembox says. but I see the files under ipod control/music
02:10:36KiwiCamphilipballew: When you mean update, do you mean place new files onto it? MP3's etc?
02:10:59philipballewI mean I used rythembox to load files onto my ipod
02:11:17KiwiCamWhat sort of files?
02:11:33saratoga_i guess he means add music rather then update rockbox
02:11:33philipballewall my music files
02:11:34JdGordonKiwiCam: the clip is a 1bpp device, so transparency is just white pixels
02:11:55philipballewI put music on my ipod
02:12:09KiwiCamJdGordon: Thanks.
02:12:37KiwiCamHave you looked at gPodder?
02:13:36philipballewsaratoga, after I put music on my ipod do i need to update rockbox to see it?
02:14:01saratoga_you just need to have the database update enabled, which i think it is by default, but double check that yours is on
02:14:27philipballewwhere do i check for that?
02:14:35 Quit ender| (Ping timeout: 240 seconds)
02:14:40saratoga_settings somewhere
02:14:58philipballewalright. ill unmount now
02:15:58 Quit hilbert (Quit: Textual IRC Client:
02:16:43 Join hilbert [0] (
02:18:04KiwiCamphilipballew: Do have a look at gpodder some time. It works well with the iPod and has some extra configuration stuff hidden in the Preferences/Edit Config and an auto sync pluggin specifically for the iPod.
02:18:10 Join ender| [0] (
02:19:09philipballewalright. ill see if I can download and try it out this evening. It looks good to me from the screenshots
02:21:18KiwiCamTry some Podcasts too I highly reccommend Dan Carlin's Hardcore History. It's the best.
02:21:21*KiwiCam is away: Oh look! What's that over there...
02:22:21saratoga_KiwiCam: be glad i can't remember how to ban your client
02:22:59 Quit Whopper|Away (Ping timeout: 240 seconds)
02:23:55philipballewwell thanks everyone!
02:24:00 Quit philipballew (Quit: Peace Out!!!)
02:24:02 Join Whopper|Away [0] (colin@about/cooking/sandwich/whopper)
02:26:58Mode"#rockbox +o scorche" by ChanServ (ChanServ@services.)
02:27:13Mode"#rockbox +q KiwiCam!*@*" by scorche (~scorche@rockbox/administrator/scorche)
02:28:03scorcheKiwiCam: let me know when you have disabled the public channel away notifications on your client and I will un-mute you
02:28:34[Saint]Did I miss a warning?
02:28:48[Saint]Or did that happen earlier...?
02:28:50saratoga_like 3 or 4 of them
02:29:00[Saint]Ah, right.
02:29:23[Saint]The concept of topics seems lost on the boy.
02:30:08[Saint]...apparently my client dropped, but, didn't. I've got lots of scrollback but no traffic over the past hours.
02:30:15 Quit hilbert (Quit: Textual IRC Client:
02:31:03*scorche pushes [Saint] to the correct channel
02:39:02[Saint]JdGordon: will, or wont the monochrome targets accept a 16/24 bit colour bitmap provided only black/white is used?
02:39:14[Saint]...that's something I'm unclear on.
02:40:28JdGordoni believe it will store it internally as a mono bitmap
02:40:36JdGordonbut im not sure and cant check right now
02:41:32LloreanOn pure mono targets it would make sense that they do at least. That's where I wasn't thinking last night, being the very early morning. There's no reason to not convert non-mono bitmaps to mono on the clip
02:41:38[Saint]Right...I can't check either and wasn't confident enough to cry "lies!" last night when a Llorean was explaining why it wouldn't work.
02:42:27LloreanOn non-mono targets, mono bitmaps are handled specially. The question is really whether there's code that specially handles non-mono bitmaps on mono-targets to store them in the more efficient way.
02:44:12[Saint]Colour targets have such magic...we only support 16bit images, but 24bit images are scattered everywhere with no cost to skin ram.
02:44:36[Saint]It'd be weird if non-colour targets didn't have similar magic.
02:45:01LloreanThe difference is that there's no case where a 24-bit image *isn't* converted
02:45:13*JdGordon bangs his head on the desk
02:45:25LloreanBut like I said, I don't know.
02:45:32JdGordonfor the umpteenth time... images make no difference to the skin ram buffer usage anymore
02:45:38LloreanLast night I was tired and not really thinking about the Clip as the special case.
02:45:46JdGordona 10MB bmp will fit in the 80kb skin buffer
02:46:27*Llorean thinks being on ignore is a headache.
02:46:41LloreanRelay to him that we get that it doesn't use the "skin buffer" but it still obviously uses some RAM somewhere.
02:48:04[Saint] does?
02:48:27LloreanI don't even get what you're asking.
02:48:37LloreanHow would you use an image in a WPS/Skin/Theme without it using *any* RAM at all?
02:48:49[Saint]"Why does it use more ram *somewhere*?"
02:49:02LloreanI didn't say "more"
02:49:20LloreanI said that even if images don't use "skin buffer" they obviously still use *some* ram somewhere.
02:49:58LloreanWhich makes the question of whether they're converted still relevant even if the skin buffer size itself isn't getting in the way.
02:50:48LloreanAs I've said a few times, I don't know whether 16-bits use more on mono targets, so me saying "it obviously uses more" would be directly contradictory now.
02:51:08[Saint]It was my understanding tthat they are still using the skin buffer, but the image is converted so that its original state doesn't matter a fuck.
02:51:33LloreanJdGordon has said a few times images don't use skin buffer.
02:51:39LloreanHe said it last night too.
02:52:41[Saint]Then saying "a 10MB bitmap will still fit in an 80k buffer" was either happily ignored or missed.
02:53:09[Saint](I have no c/p in this client...its a few lines up)
02:53:23LloreanDo you honestly believe he means that a 10MB bitmap is converted to a format that fits in less than 80k?
02:53:37LloreanHe was saying that the bitmap isn't stored on the buffer, so even though the buffer is too small it won't restrict the use of such an image.
02:53:43LloreanRead the rest of his statements for context.
02:54:49Llorean[Saint]: (3:50:43 AM) JdGordon: Kiwi_CAM: the images dont load into the skin buffer so their size makes no difference at all
02:55:00LloreanI don't think it can be any more explicit than that.
03:00:19Unhelpfulanybody could've looked at the loader to see what is converted, and how. unless somebody has changed it,iirc, loading for skins uses a static buffer, never ever touches the scaler, and is converted to target-native bitmap always - many targets have framebuffer formats that in no way resemble any possible bmp format. non-mono (or only color?) targets can request a bitmap be loaded as mono.
03:05:09LloreanUnhelpful: So on color targets, when a bitmap is loaded as mono, is it still stored in the framebuffer format or is that an exception?
03:05:54Unhelpfulpretty sure it's stored in the "mono" format then... but i haven't looked in a very long time, and the only thing i am certain of is that it is #ifdef hell.
03:06:03LloreanOkay, that's what I thought and was trying to express
03:06:13LloreanThe part where I screwed up was not making the leap "framebuffer = mono on clip"
03:21:39 Join simonlnu [0] (3WO7PTYvjG@unaffiliated/simonrvn)
03:40:18 Quit [Saint] (Quit: flashing...will I be back? Depends if the rom plays nice.)
03:47:33 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
04:00:21***Saving seen data "./dancer.seen"
04:05:35freddybDoes anyone have any objection to FS #12333 - Faster cached font loading? I'd really like to commit it soon.
04:05:36fs-bluebot Faster cached font loading (patches, new)
04:09:18 Quit Scromple (Read error: Connection reset by peer)
04:14:11[Saint]Hum...interesting patch freddyb, I should watch the tracker more closely.
04:14:38 Quit amiconn (Disconnected by services)
04:14:39 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:14:39 Quit pixelma (Disconnected by services)
04:14:42 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:14:44 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:15:06 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:15:39 Join Scromple [0] (
04:16:02 Join kadoban_ [0] (
04:17:50 Quit kadoban (Ping timeout: 258 seconds)
04:18:55freddyb[Saint]: Thanks. I'm getting fonts to load 2-6 times faster.
04:19:32[Saint]Test targets?
04:20:29freddybIpod Video
04:21:16freddybIt's an optimization for hard disks.
04:24:22[Saint]Right...I know that much. Just wondering if some disks showed more improvement than others.
04:24:43freddybI'll try on a Fuzev1 here in a minute
04:30:40 Quit TheSeven (Disconnected by services)
04:30:56 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
04:34:49freddybWell, it's about 4-5 times faster than stock on a Fuze v1 with 16-GNU-Unifont
05:05:21 Quit [Saint] (Read error: Connection reset by peer)
05:05:41 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
05:06:11 Quit robin0800 (Quit: Leaving)
05:06:40 Join robin0800 [0] (~robin0800@
05:22:14 Quit freddyb (Quit: Leaving)
05:34:25 Quit Horscht (Quit: Verlassend)
05:48:23 Quit kadoban_ (Remote host closed the connection)
05:48:44 Join kadoban_ [0] (
05:48:58 Quit kadoban_ (Remote host closed the connection)
05:49:47 Join kadoban [0] (
05:51:03 Join FBI_Guy [0] (~fbiguy@
05:53:11FBI_Guyanyone know wgere the latest dfkt builds reside?
05:54:31 Quit FBI_Guy (Client Quit)
05:58:07[Saint]dfkt builds?
05:58:37 Join Rob2223 [0] (
06:00:24***Saving seen data "./dancer.seen"
06:01:54 Quit Rob2222 (Ping timeout: 252 seconds)
06:16:25 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
06:19:57 Quit [Saint] (Ping timeout: 260 seconds)
06:20:53 Join ChickeNES [0] (~ChickeNES@
06:28:37 Quit ChickeNES (Read error: Connection reset by peer)
06:28:58 Join ChickeNES [0] (~ChickeNES@
06:30:24 Join Buschel [0] (
06:31:22 Quit robin0800 (Quit: Leaving)
06:35:49 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
06:37:50 Quit [Saint_] (Ping timeout: 252 seconds)
06:58:10 Quit ChickeNES (Quit: Computer has gone to sleep.)
07:04:50scorcheKiwiCam: see my private query
07:06:42 Join Keripo [0] (
07:09:44Mode"#rockbox -q KiwiCam!*@*" by scorche (~scorche@rockbox/administrator/scorche)
07:09:59Mode"#rockbox -o scorche" by ChanServ (ChanServ@services.)
07:14:39KiwiCamThen there's this comment " JdGordon: Kiwi_CAM: the images dont load into the skin buffer so their size makes no difference at all" but what about this one? "<JdGordon> it is just generic storage for the skins... the more tags you use (and viewports, etc) the more you will use". All I'm trying to d, is find out what's stopping me from building bigger themes. So, if the images aren't in the Skin Buffer, then JdGordan clarified that fo
07:14:39KiwiCamr me by saying that it's the Tags and Viewports. Therefore it must be the limitation on the Viewports I need cause because of the Skin Buffer. Or, am I just talking rubbish?
07:16:56[Saint]There's also a MAX_TOKEN (I _think_) limit you can hit, without exceeding the skin buffer.
07:17:13[Saint]Hitting that will cause things to not load.
07:18:09[Saint]This is the limit on declarations a skin can make, ie, skin tags.
07:18:38[Saint]Its set to a reasonable sane level, but, I've overflowed it before with extravagant themes.
07:19:37JdGordon[Saint]: no, that value is a soft limit to size the buffer
07:19:52JdGordonthe only hard limit in the entire skin system is the buffer size
07:20:26KiwiCamSorry Guys.I'd like to apologise again. When I read point 7, I assumed it was something I was doing wrong with my username (the second part of point 7). I didn't realise I was running a script.
07:20:36JdGordonevery single element in the skin allocates 64 bytes (err, maybe more)
07:21:48 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238])
07:22:39KiwiCam[Saint]: MAX-TOKEN, limits on declarations. These are terms that I'm not familiar with. I'm wondering if this explains why I can do more in the Simulator than with the actual Clip. Is the simulator using a more recent build perhaps? ie one that includes the increased buffer size?
07:23:03JdGordonKiwiCam: I suggest you make a theme as big as you can, if it works on the sim but not on target then take it up with you-know-who
07:23:30JdGordonor, get the compilers going and build your own build
07:23:45KiwiCamErr....who would that be? :(
07:24:39[Saint]In this case, kugel, KiwiCam .
07:24:49[Saint]I would expect so anyway.
07:25:19KiwiCamI wouldn't be able to share my themes then, if I had my own build. There'd be no fun in that.
07:25:57[Saint]You could share your build and sources...
07:25:59KiwiCamAnd Kugel maintains the Clip builds (svn's?)
07:27:23KiwiCam(BTW: It's a darned good thing that I left Xchat running on this machine. Otherwise, I'd have NEVER seen these comments about my misbehaviour)
07:31:35 Quit KiwiCam (Quit: Leaving)
07:31:54 Join KiwiCam [0] (
07:33:29 Quit saratoga_ (Quit: Page closed)
07:45:47 Quit [Saint] (Quit: AndChat - Android IRC Client -
07:48:52 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
07:49:52kugelpahh, JdGordon still has and uses flyspray powers
07:54:28 Quit KiwiCam (Quit: Leaving)
07:54:45 Join KiwiCam [0] (
08:00:28***Saving seen data "./dancer.seen"
08:09:44 Join Kiwi_CAM [0] (
08:13:53 Join GodEater_ [0] (93722cc8@rockbox/staff/GodEater)
08:17:18 Quit Keripo (Ping timeout: 248 seconds)
08:21:24 Quit beta2k (Ping timeout: 240 seconds)
08:21:34 Join beta2k [0] (
08:32:47 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:35:58 Join nick-p [0] (
08:36:32 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
08:43:35 Join einhirn [0] (
08:47:48 Join bertrik [0] (
08:47:49 Quit bertrik (Changing host)
08:47:49 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:50:49 Quit Zarggg (Quit: Rebooting client...)
08:53:00 Join ender` [0] (
09:04:20 Join JdGord [0] (
09:05:00 Quit antil33t (Read error: Connection reset by peer)
09:05:10 Join antil33t [0] (
09:06:51 Join ender [0] (
09:07:02 Quit ukleinek (Ping timeout: 244 seconds)
09:07:08 Quit amiconn (Disconnected by services)
09:07:10 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
09:07:32 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
09:07:40 Quit ender` (Read error: Connection reset by peer)
09:07:46 Quit [7] (Quit: No Ping reply in 90 seconds.)
09:08:05 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
09:09:18 Quit Scromple (Read error: Connection reset by peer)
09:12:09 Quit JdGord (Ping timeout: 276 seconds)
09:17:50 Join hilbert [0] (
09:19:12 Join JdGord [0] (~AndChat@
09:24:30 Quit JdGord (Ping timeout: 276 seconds)
09:26:50 Join pamaury [0] (
09:26:50 Quit pamaury (Changing host)
09:26:50 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:29:06 Quit kadoban (Ping timeout: 245 seconds)
09:31:07 Join petur [0] (~petur@rockbox/developer/petur)
09:32:54 Quit bertrik (Read error: Operation timed out)
09:37:43 Join markun [0] (~markun@rockbox/developer/markun)
09:42:52 Join LinusN [0] (
10:00:29***Saving seen data "./dancer.seen"
11:07:31 Join wodz [0] (
11:09:52wodzWhy pcm_dma_apply_settings() is named like this? From what I see it sets codec/i2s samplerate and has nothing to do with DMA settings
11:13:06wodzjhMikeS: ^
11:14:14kugelwodz: pcm_dma_* is generally the target specific code, only remotely related to actual dma
11:17:40wodzkugel: the name is misleading
11:18:22Zagortoday is 3.10 feature freeze
11:19:39wodzZagor: I would like to commit hifiman 60x port (FS #12319) but AFAIK this doesn't count as this will not be stable target
11:19:40fs-bluebot HiFiMAN HM-601 initial work (patches, unconfirmed)
11:20:12Zagorwodz: right. unless it modified a lot of other code.
11:21:13kugelZagor: huh?
11:21:19wodzonly rk27xx specific
11:21:40kugelcan I still commit my patches from the ML? (I was barely able to commit the last days due to lack of laptop)
11:22:11Zagorwodz: yeah, that patch looks fine to me
11:22:22Zagorkugel: well we haven't officially called freeze yet :)
11:22:28Zagor(on the mailing list)
11:22:43wodzbtw what does unconfirmed mean in fs-bluebot output?
11:22:59Zagorwodz: "fine" as in "doesn't voilate freeze". I haven't actually looked at the code
11:23:23Zagorwodz: it's a flyspray status field, intended for bugs
11:23:24kugelright. but I can't do anything until 6pm (cest). I'm at work in this moment
11:24:16wodzZagor: sorry I don't understand - what constitutes 'voilate freeze' than?
11:24:55Zagorwodz: spelling error- "violate freeze" would be a patch that introduces new features while in feature freeze
11:26:17wodznew feature you mean by affecting release targets I guess
11:26:35wodzthan this FS entry doesn't count
11:27:32JdGordonwanna commit fs#12328 before the free starts?
11:27:33fs-bluebot move icons onto buflib and remove the 24x24 max iconsize limitation (patches, new)
11:28:31kugelZagor: is this extended freeze (as people thought about earlier) or the normal 7 days?
11:29:05gevaertskugel: I'd say that depends on how stable things are one week from now
11:29:08Zagorthis is the freeze as planned in the rockbox calendar.
11:29:45*gevaerts isn't at all convinced that moving more stuff to buflib *now* is a good idea
11:29:53Zagorif we manage to get it releaseable on schedule, we won't have to decide about length :)
11:30:04JdGordonwell if it is "unstabe" anyway, may as well
11:30:17Zagorgevaerts: what patch does that?
11:30:28gevaertsZagor: the one JdGordon mentioned
11:30:42Zagoroh, missed that :)
11:31:55ZagorI'd tentatively agree to err on the side of caution right now. we're not in a very good state.
11:32:26wodzI would like to ask for advice. lcds can be differently wired to rk27xx. The most common case would be 16bit and 18bit interface I guess (but 8bit is also possible). The way lcd is wired affects how to do bit shifts in driver.
11:32:30JdGordonI'm not sure I agree with the suggestion things are currently flakey
11:33:07wodzNow the same type of lcd can be wired differently on different targets although this is not the case right now
11:34:28Zagorwodz: why not move lcd_data_transform() into hm60x/lcd-hm60x.c then?
11:35:07Zagorand other code that depends on model-specific wiring
11:35:34Zagoror #ifdef it if splitting causes too much duplication
11:35:39wodzZagor: It is quite likely to have target with 16bit interface in the future - this will add code duplication
11:36:06Zagorcan't you just duplicate the functions that differ?
11:36:13wodzI can
11:36:51Zagoras far as I can tell from a quick glance, that wouldn't create very much code duplication?
11:37:01wodzbut we will duplicate the same function over and over if more targets appear
11:37:30wodz*one from the two variants
11:37:42Zagorthis is not carved in stone. if >1 target have the same interface, we can always do it another way.
11:39:14 Join [Saint] [0] (65628198@gateway/web/freenode/ip.
11:39:30pamaurywodz: if I were you I would just wait for new targets before changing things
11:40:00Zagoryeah, don't plan too much ahead. make it simple now.
11:40:18Zagorin two years, perhaps another solution will be simpler. we'll do that then. not now.
11:40:35wodzyou are right probably :-)
11:41:13pamaurythat's the same thing for me, I tried to make the imx233 code as generic as possible but only when there will be another imx233 port I will know if it's good enough or not :)
11:44:37pamauryI have one question regarding audio, perhaps other people have encountered it: when sending audio data directly to the fifo (in pio mode) on the fuze+, it plays it but then instead of muting there is a constant noise like if the last sample was indefinitely played
11:46:00 Quit fs-bluebot (Ping timeout: 260 seconds)
11:47:14 Join fs-bluebot [0] (
11:47:34 Quit bluebrother (Ping timeout: 260 seconds)
11:48:32wodzpamaury: you mean whe you stop sending samples?
11:49:04 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
11:54:24Kiwi_CAMVanity question: If you update a theme, do you lose your Votes (ratings) like you used to?
11:58:05[Saint]It makes sense, too.
11:58:23[Saint]people voted for the older revision..the newer revision could suck balls.
11:58:59wodzpamaury: I had similar result when I implemented pcm_play_dma_pause() by gating clock to i2s module. Now I disable clock to dma and this seems to work.
11:59:42Kiwi_CAMI hadn't thought about it that way. Sounds fine by me.
11:59:48pamauryI can't make dma works for now, it's just with pio
12:00:30***Saving seen data "./dancer.seen"
12:01:03[Saint]Kiwi_CAM: I used to get grumpy about it, one of my themes used to have an *excellent* rating (and probably does again) based on several hundred (almost a thousand) votes
12:01:11[Saint]...then, I got over it ;)
12:09:30 Quit Kiwi_CAM (Quit: Leaving)
12:12:14 Join JdGord [0] (
12:32:21CIA-14New commit by wodz (r30765): Add HiFiMAN HM-60x target(s). FS #12319 by Andrew Ryabinin with some (small) modification by me. This also splits rk27xx lcd driver into lcdif-rk27xx ...
12:35:17CIA-14r30765 build result: 0 errors, 30 warnings (wodz committed)
12:37:02wodzthis is pure magic for me - how this warning is only on some buildclients?
12:38:37CIA-14New commit by wodz (r30766): fix yellow
12:40:06amiconnProbably depends on the version of 'make'
12:41:08CIA-14r30766 build result: All green
12:44:12CIA-14New commit by wodz (r30767): Credit Andrew Ryabinin for hifiman-hm60x port
12:45:15wodzZagor: I guess we would like to welcome him as well as maintainer for this work (aka svn access)
12:46:11CIA-14r30767 build result: All green
12:58:19 Quit kugel (Quit: leaving)
12:59:07 Quit wodz (Quit: Leaving)
13:01:08 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:01:22 Quit kugel (Client Quit)
13:01:25 Quit nick-p (Quit: Leaving)
13:03:37 Join kugel [0] (~kugel@
13:03:38 Quit kugel (Changing host)
13:03:38 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:25:41 Quit kugel (Quit: leaving)
13:26:10 Join kugel [0] (~kugel@
13:26:10 Quit kugel (Changing host)
13:26:10 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:57:38 Join T44 [0] (
14:00:33***Saving seen data "./dancer.seen"
14:01:18 Quit Topy (Ping timeout: 248 seconds)
14:11:01 Join MethoS- [0] (~clemens@
14:11:24CIA-14New commit by fredwbauer (r30768): Update UI simulator buttonmap for Sansa Fuze
14:13:29CIA-14r30768 build result: All green
14:15:55 Join jdgord_ [0] (~AndChat@
14:16:58 Quit jdgord_ (Client Quit)
14:18:21JdGordonZagor: Torne: any updates on git? :)
14:18:54 Quit JdGord (Ping timeout: 248 seconds)
14:19:43ZagorI'm waiting for Torne
14:19:53Zagorhe has the ball at the moment
14:25:09 Join nick-p [0] (
14:28:49 Join freddyb [0] (
14:29:05 Quit factor (Quit: Leaving)
14:39:42 Join Topy [0] (
14:43:26 Quit T44 (Ping timeout: 248 seconds)
14:52:29TorneYeah, will get there
14:52:46TorneI'm taking a week off work to work on personal stuff soonish :)
15:00:38 Join TheLemonMan [0] (
15:04:11 Quit antil33t (Read error: Connection reset by peer)
15:04:26 Join matze` [0] (
15:04:35 Join antil33t [0] (
15:08:00TorneZagor: but first i have to finish a process at work we are referring to as the "terror merge" :)
15:17:01 Quit hilbert (Ping timeout: 245 seconds)
15:17:05CIA-14New commit by fredwbauer (r30769): Faster cached font loading. FS #12333
15:18:08 Join hilbert [0] (
15:19:30CIA-14r30769 build result: 0 errors, 46 warnings (fredwbauer committed)
15:19:55 Join dfkt [0] (
15:19:56 Quit dfkt (Changing host)
15:19:56 Join dfkt [0] (dfkt@unaffiliated/dfkt)
15:25:34 Join robin0800 [0] (
15:29:17CIA-14New commit by fredwbauer (r30770): Split long fmt string which caused a yellow
15:29:21 Join Zarggg [0] (
15:31:19CIA-14r30770 build result: 0 errors, 1 warnings (fredwbauer committed)
15:36:49 Join swilde [0] (
15:38:43 Join WalkGood [0] (~4@unaffiliated/walkgood)
15:45:34jhMikeSwodz: maybe misleading a bit :-) but it was merely meant to be consistent with expectations of being target code
16:00:34***Saving seen data "./dancer.seen"
16:08:11 Join mortalis [0] (
16:16:27 Part nick-p ("Leaving")
16:21:35 Join T44 [0] (
16:24:46 Quit Topy (Ping timeout: 248 seconds)
16:25:34 Quit mortalis (Quit: CGI:IRC)
16:40:01 Quit robin0800 (Quit: Leaving)
16:45:36 Quit [Saint] (Quit: Page closed)
16:52:37 Join mortalis [0] (~mortalis@
16:53:31 Join webguest35 [0] (
16:53:53 Quit TheLemonMan (Quit: WeeChat 0.3.5)
17:00:40 Quit webguest35 (Quit: CGI:IRC)
17:00:50 Join webguest46 [0] (
17:01:36 Part Zagor
17:02:21mortaliscould someone give me access on wiki?
17:02:41mc2739mortalis: yes, what is your wiki name?
17:02:56mortalismc2739: AndrewRyabinin
17:04:28 Join Pogglywoggly [0] (
17:04:56 Part Pogglywoggly
17:06:26 Quit webguest46 (Quit: CGI:IRC)
17:07:24 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:08:40 Quit WalkGood (Quit: CaminaBien ♪ ♫ ♪ ♫)
17:12:04 Quit GodEater_ (Quit: Page closed)
17:32:24 Join n1s [0] (~quassel@rockbox/developer/n1s)
17:37:16CIA-14New commit by jethead71 (r30771): Gigabeat S/i.MX31/wm8978: Clean up clocking information in the general wmcodec/pcm drivers and move it to the target's wmcodec/i2s files.
17:39:29CIA-14r30771 build result: All green
17:40:39 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
17:48:42CIA-14New commit by jethead71 (r30772): Expunge a commented #include that found its way in in r30771.
17:49:50 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
17:50:17 Quit hilbert (Quit: Textual IRC Client:
17:50:26 Join kadoban [0] (
17:50:44CIA-14r30772 build result: All green
18:00:36***Saving seen data "./dancer.seen"
18:04:26 Join y4n [0] (y4n@unaffiliated/y4ndexx)
18:13:12 Join robin0800 [0] (
18:16:53 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:16:54 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:17:32 Quit petur (Quit: *plop*)
18:18:41 Quit n1s (Ping timeout: 245 seconds)
18:23:46 Join hilbert [0] (
18:44:55*pamaury has another try at fuze+ audio dma
18:49:39 Quit swilde (Remote host closed the connection)
19:00:17 Join TheLemonMan [0] (
19:03:53 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
19:08:21pamauryhum, *not* setting the dac frequency doesn't help playing audio...
19:10:58 Join dfkt [0] (dfkt@unaffiliated/dfkt)
19:13:27 Quit simonlnu (Quit: brb)
19:17:16 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
19:17:53 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
19:21:04 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
19:34:46 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:37:07bertrikpamaury, can I help with something?
19:37:43 Quit robin0800 (Quit: Leaving)
19:37:44bertrikon AMS we send a 128xFs clock to the DAC (MCLK) and a clock signal at the sample rate (LRCLK)
19:38:12CIA-14New commit by kugel (r30773): Commit FS #12321 - Touchscreen: List line padding, to more easily select lines ...
19:38:31 Quit kadoban (Remote host closed the connection)
19:39:00pamaurybertrik: no, currently I don't really understand why it doesn't work, must be missing some init
19:39:15 Quit benedikt93 (Quit: Bye ;))
19:40:08pamauryI'm getting closer but still I only get garbage and dma never send data
19:40:24pamauryor I forgot some interrupt enable somwhere
19:40:40CIA-14r30773 build result: 175 errors, 197 warnings (kugel committed)
19:41:38 Quit TheLemonMan (Quit: WeeChat 0.3.5)
19:42:01 Join Buschel [0] (
19:43:03bertrikdma is sometimes one of those things that looks very easy when it works, but it not easy to get working
19:45:05bertrikkugel, does that commit do something for non-touchscreen targets too?
19:49:21CIA-14New commit by kugel (r30774): Fix checkwps reds and non-touchscreen warnings.
19:51:15CIA-14r30774 build result: 24 errors, 21 warnings (kugel committed)
19:53:38pamaurybertrik: I know dma works since I already have a general framework which works for ssp and i2c
19:55:04pamaurybut to have working audio you have to enable lots of things on this chip...
19:59:03 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
20:00:40***Saving seen data "./dancer.seen"
20:01:19 Quit y4n (Quit: PÆNTS ØLF!)
20:01:45 Join y4n [0] (y4n@unaffiliated/y4ndexx)
20:03:34 Join petur [0] (~petur@rockbox/developer/petur)
20:07:13 Quit hilbert (Quit: Computer has gone to sleep.)
20:08:57 Join hilbert [0] (
20:13:02 Quit Buschel (Ping timeout: 248 seconds)
20:28:38CIA-14New commit by kugel (r30775): - Fix some reds by implementing lcd_get_dpi(). ...
20:29:54 Join simonlnu [0] (zug5etmxap@unaffiliated/simonrvn)
20:30:47CIA-14r30775 build result: All green
20:33:29 Quit mortalis (Quit: KVIrc 4.1.1 Equilibrium
20:37:32 Join Keripo [0] (
20:42:10 Quit Keripo (Ping timeout: 255 seconds)
20:46:34 Join dfkt [0] (dfkt@unaffiliated/dfkt)
20:54:01 Quit y4n (Read error: Connection reset by peer)
20:56:04 Join PokeYouInDaeye [0] (
20:56:21 Quit PokeYouInDaeye (Client Quit)
20:56:44 Join PokeYouInDaeye [0] (
20:57:09PokeYouInDaeyeanyone around familiar with installations?
20:57:35CIA-14New commit by kugel (r30776): Move Time & Date menu from System to settings.
20:57:40CIA-14New commit by kugel (r30777): Sleep timer options: persistent duration and start on boot. ...
20:57:46CIA-14New commit by kugel (r30778): Remake the sleep timer menu item, so that selecting it while the timer is running ...
20:58:01PokeYouInDaeyeI'm wondering if, having a version of Rockbox on myPod Classic 6g already, can I just download the newer .zip file and replace the /.rockbox/ directory with the new .zip file-extraction?
20:59:08bertrikThe answer is yes
20:59:39PokeYouInDaeyefor me bertrik?
20:59:41PokeYouInDaeyethanks ddue
20:59:42CIA-14r30776 build result: All green
21:02:05CIA-14r30778 build result: All green
21:02:38 Quit avacore (Ping timeout: 248 seconds)
21:04:12 Join avacore [0] (
21:06:20PokeYouInDaeyeI rub rockbox.
21:06:33PokeYouInDaeyeno longer just an iPod... it is now myPod
21:06:33 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
21:07:51dfktdoes anyone know where to get the "shutdown in main menu" patch for android? the dropbox location is 404ed:,27394.msg184533.html#msg184533
21:09:05 Quit PokeYouInDaeye (Quit: CGI:IRC (EOF))
21:09:30amiconnkugel: Quite large delta for what it does :\
21:10:39amiconnAlsoo seconds_to_min can be significantly simplified. No need to do two costly divisions
21:11:08amiconnit becomes just: return (secs + 10) / 60;
21:13:58amiconnAlso why do you have (void)parameter; for parameters which are actively used?
21:17:01 Join TheLemonMan [0] (
21:21:27CIA-14New commit by amiconn (r30779): Simplify seconds_to_min(), and don't safeguard used parameter.
21:23:30CIA-14r30779 build result: All green
21:24:27dfktthank you, mc2739
21:25:33kugelamiconn: thanks
21:35:58 Join mudd1 [0] (
21:44:05 Join y4n [0] (y4n@unaffiliated/y4ndexx)
21:51:35 Join z180 [0] (
22:00:42***Saving seen data "./dancer.seen"
22:11:12 Quit z180 (Ping timeout: 260 seconds)
22:12:27 Join webguest58 [0] (
22:12:42 Quit webguest58 (Client Quit)
22:21:26 Quit ack` (Remote host closed the connection)
22:26:33 Quit petur (Remote host closed the connection)
22:29:42 Join Strife89 [0] (~Strife89@
22:31:27CIA-14New commit by wodz (r30780): Add HiFiMAN HM-601/HM-602
22:31:50 Join wodz [0] (
22:32:18 Quit TheLemonMan (Quit: WeeChat 0.3.5)
22:33:52wodzBagder: could you update frontpage please?
22:34:18 Quit thomasjfox (Remote host closed the connection)
22:44:27 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
22:49:06 Join Keripo [0] (
22:51:00Zagorkugel: why the menu.h change in 30778?
22:53:00 Join Jerom [0] (~jerome@
23:00:56 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
23:08:03ZagorI don't understand why that change was in that commit. or how 30776 could compile with it but 30778 can't.
23:09:29 Join ack [0] (
23:11:28 Nick ack is now known as ack` (
23:28:54*pamaury goes for some dma debugging...
23:30:52 Join nick-p [0] (
23:32:25 Join liar [0] (
23:37:31 Quit nick-p (Quit: Leaving)
23:38:12 Quit domonoky (Read error: Connection reset by peer)
23:41:15Zagor-Os enables no less than 27 different optimization steps that -O does not use. (arm-elf-eabi-gcc v4.4.4)
23:42:08Zagor"arm-elf-eabi-gcc -c -Q -O −−help=optimizers" lists all optimizer steps
23:42:55Zagorthat shows -O2 is the same as -Os except Os has -finline-functions *enabled*!
23:43:43Zagorthis suggests we should use -Os on all targets
23:46:37 Quit Strife89 (Quit: Heading out.)
23:46:46Zagorit definitely saves a ton of space
23:51:03 Quit ender (Quit: Intelligence is the ability to avoid doing work, yet getting the work done. -- Linus Torvalds)

Previous day | Next day