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 2011-11-30

00:02:23JdGordongeneric pong
00:02:44JdGordonsaratoga_: i tihnk rasher might need to manually fix the database
00:02:48saratoga_ah ok
00:06:48 Quit fyrestorm (Read error: Connection reset by peer)
00:11:06saratoga_i've posted proposed link changes for the front page in FS #12416
00:11:06fs-bluebot Minor Front Page tweaks (patches, new)
00:11:11saratoga_comments are welcome
00:15:59 Join Scr0mple [0] (
00:16:21 Quit bertrik (Quit: And That, My Liege, Is How We Know the Earth to Be Banana Shaped)
00:18:38 Quit Scromple (Ping timeout: 248 seconds)
00:22:17 Quit pamaury (Remote host closed the connection)
00:22:54 Quit domonoky (Ping timeout: 248 seconds)
00:23:41 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)
00:29:35 Quit MethoS- (Read error: Connection reset by peer)
00:29:40 Join MethoS-- [0] (~clemens@
00:33:13 Quit lebellium (Quit: ChatZilla 0.9.87 [Firefox 9.0/20111122192043])
00:37:37 Quit MethoS-- (Remote host closed the connection)
00:37:53 Join MethoS-- [0] (~clemens@
00:54:36 Join Scromple [0] (
00:57:02 Quit Scr0mple (Ping timeout: 248 seconds)
00:57:51JdGordondoes doom resize to fit the screen?
01:00:14 Join dreamlayers [0] (
01:00:14 Quit dreamlayers (Changing host)
01:00:14 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
01:01:38saratoga_i think so
01:06:31***Saving seen data "./dancer.seen"
01:28:22 Quit Strife89 (Quit: Heading out.)
01:33:49 Join Scr0mple [0] (
01:35:42 Quit Scromple (Ping timeout: 252 seconds)
01:37:34 Join Scromple [0] (
01:38:38 Quit Scr0mple (Ping timeout: 248 seconds)
01:39:57 Join Topy44 [0] (
01:40:46 Quit T44 (Ping timeout: 248 seconds)
01:51:08 Quit ReimuHakurei (Quit: Leaving...)
01:51:08 Nick Reimu_Hakurei is now known as ReimuHakurei (
01:54:57 Quit NoPinky ()
02:05:57 Quit Thra11 (Quit: kthxbai)
02:13:00 Join Keripo [0] (
02:24:33 Quit dreamlayers (Ping timeout: 244 seconds)
02:25:35 Quit hilbert (Quit: Textual IRC Client:
02:31:12 Join ReimuHak_ [0] (
02:35:44 Quit Keripo (Quit: Leaving.)
02:42:36 Join Keripo [0] (
02:43:10 Quit [sanyi] (Ping timeout: 248 seconds)
02:52:32 Join [sanyi] [0] (~sanyi@unaffiliated/alexanderc)
02:56:51 Join mordocai [0] (
03:01:45 Join kadoban [0] (
03:06:34***Saving seen data "./dancer.seen"
03:12:57 Quit Keripo (Ping timeout: 245 seconds)
03:15:24 Quit MethoS-- (Remote host closed the connection)
03:27:18 Quit mordocai (Quit: reboot for kernel updates)
03:52:10 Join wawaron [0] (
03:52:34 Part wawaron
04:06:43CIA-10New commit by mc2739 (r31096): Text viewer plugin: Fix info viewport font not being set to user selected font
04:08:37CIA-10r31096 build result: 2 errors, 0 warnings (mc2739 committed)
04:12:49CIA-10New commit by mc2739 (r31097): Backport r31096 - Text viewer plugin: Fix info viewport font not being set to user selected font
04:26:37 Quit [7] (Disconnected by services)
04:26:47 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:31:01 Join dreamlayers [0] (
04:31:02 Quit dreamlayers (Changing host)
04:31:02 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
04:45:10 Quit amiconn (Disconnected by services)
04:45:11 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:45:13 Quit pixelma (Disconnected by services)
04:45:15 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:45:17 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:45:33 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:50:30dreamlayersWhat is the difference between firmware/include and firmware/export?
04:55:56JdGordonisnt include more for libc type includes?
04:58:55dreamlayersThere's also firmware/libc/include/, and there's firmware/include/font_cache.h, which definitely isn't libc type stuff.
04:59:46dreamlayersI used to think include was for within the firmware, and export was for exporting to code outside it, but that's not it either.
05:00:18JdGordonyeah, its very much "do as you please" territory now
05:02:02dreamlayersOne significance of firmware/libc/include is that it's not included when building a sim.
05:06:19JdGordonwhy do we even still have the sim builds as a sperate build?
05:06:38***Saving seen data "./dancer.seen"
05:06:41JdGordonIt seems that the "sim" build shuold really just be a sdl-app build preconfigured with different features and screen size
05:07:52dreamlayersThere are some specifics not used in the app, like remotes and the character-based LCD on the Archos Player.
05:09:55dreamlayersBesides such things, and running from simdisk, I don't think there's much difference between sim and app builds.
05:52:21 Join Rob2223 [0] (
05:55:52 Quit Rob2222 (Ping timeout: 252 seconds)
06:03:56 Quit dreamlayers (Quit: Goodnight)
06:11:56 Join funman [0] (~fun@rockbox/developer/funman)
06:21:34 Join dartmouth [0] (
06:38:49 Join jgsprenger [0] (
06:39:21jgsprengerhi gang... where can i pick up 310RC0 for e200v1?
06:39:55JdGordoncheck the forums
06:41:30jgsprengerumm ok... thought there was a short cut link somewhere handy...oh well
06:43:30 Quit jgsprenger (Client Quit)
07:06:42***Saving seen data "./dancer.seen"
07:13:00 Join HaloNachos117 [0] (
07:14:15 Quit HaloNachos117 (Client Quit)
07:32:44 Quit amiconn (Disconnected by services)
07:32:44 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
07:32:52 Quit pixelma (Disconnected by services)
07:32:54 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
07:32:56 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
07:33:06 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
07:42:26 Join HaloNachos117 [0] (
07:54:12 Quit kadoban (Ping timeout: 240 seconds)
07:54:46HaloNachos117Hello, first off, I should say I'm new to IRC. I have a question about the Fuze+ port: How does one lock the Fuze+, so that they don't accidently push buttons while the device it in their pocket. I can't seem to figure this out, although others on the forum have done it. Please help.
07:58:00 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:05:57 Quit amiconn (Remote host closed the connection)
08:05:57 Quit pixelma (Remote host closed the connection)
08:07:01 Join amiconn [0] (quassel@rockbox/developer/amiconn)
08:07:03 Join pixelma [0] (quassel@rockbox/staff/pixelma)
08:10:31 Quit factor (Ping timeout: 255 seconds)
08:11:24 Quit dartmouth (Ping timeout: 240 seconds)
08:13:13 Join GodEater_ [0] (93722cc9@rockbox/staff/GodEater)
08:25:51 Join factor [0] (~factor@
08:31:50 Quit Scromple (Read error: Connection reset by peer)
08:33:25JdGordonsooo.... any thoughts on automatic translation for skins? use user would have a single file translations.txt which just lists any they'd want to happen, then any plain string in a skin would check if it is there or an inbuilt translation and use it automagically?
08:33:39JdGordonignoring leading and trailing spaces
08:33:48 Quit HaloNachos117 (Ping timeout: 240 seconds)
08:35:33 Join HaloNachos117 [0] (
08:52:36 Join ender` [0] (
08:53:35 Join bertrik [0] (
08:53:35 Quit bertrik (Changing host)
08:53:35 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:57:45 Join wodz [0] (
08:58:09wodzJdGordon: are you going to commit m3u handling fix to 3.10 branch also?
09:06:33JdGordonno, but if you want go you can
09:06:44***Saving seen data "./dancer.seen"
09:08:05wodzI don't have branch checkout
09:16:25 Join einhirn [0] (
09:22:39 Join hilbert [0] (
09:33:01 Quit HaloNachos117 (Quit: No Ping reply in 180 seconds.)
09:33:37 Join HaloNachos117 [0] (
09:36:31 Quit bertrik (Ping timeout: 248 seconds)
09:44:37 Quit HaloNachos117 (Quit: No Ping reply in 180 seconds.)
09:45:14 Join HaloNachos117 [0] (
09:57:45 Quit HaloNachos117 (Remote host closed the connection)
10:01:24 Join pamaury [0] (
10:01:24 Quit pamaury (Changing host)
10:01:24 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:18:16 Quit GodEater (Ping timeout: 244 seconds)
10:20:26 Quit perrikwp (Read error: Operation timed out)
10:20:53 Join perrikwp [0] (
10:32:31 Join nosa [0] (
10:34:47 Quit nosa-j (Ping timeout: 260 seconds)
10:34:47 Nick nosa is now known as nosa-j (
10:46:51wodzfunman: ping
10:57:28JdGordon\o/ i figured out why the skin image layer patch wasnt working
10:57:35JdGordoni have no idea how to correctly fix it though
10:57:42JdGordonstupid images drawing too often :/
11:06:47***Saving seen data "./dancer.seen"
11:08:08JdGordongrr.... I got scrolling over the image working, but now a black background for the rest of the viewport instead of the backgronud image
11:08:57JdGordonAND fixed :)
11:09:15 Join nick-p [0] (
11:09:26JdGordon[saint]! come back online!!!!
11:11:06wodzJdGordon: what is the overhead of this?
11:11:06 Quit factor (Read error: Connection reset by peer)
11:11:29JdGordonpretty much 0
11:11:37wodzand ram usage impact?
11:11:49JdGordonextra buffer comes from buflib only if the skin calls for it
11:11:49 Join T44 [0] (
11:12:01 Quit Topy44 (Read error: Connection reset by peer)
11:12:19wodzso you don't allocate additional framebuffers for mixing?
11:13:32JdGordoncorrect. The extra buffer is technically the backdrop buffer for the screen
11:13:47JdGordonthe user will need to load images to draw into it, but they dont have to be the full screen size
11:14:02JdGordonand ~0 waste if your theme doesnt do this
11:15:23wodzok so this is basically the same as flatting wps elements onto backdrop but with the twist
11:15:46JdGordonthats exactly what this is
11:18:25wodzare there other big benefits despite being cool and allow scroll text over images?
11:19:42JdGordonthat's not enough? :)
11:20:24wodzDid I say something like this? :-)
11:20:36JdGordonthis could allow things like dma to speed up fullscreen updates IIUC
11:21:15kugelhow that?
11:21:24wodzyou mean dma to framebuffer or dma to lcd controller?
11:21:44JdGordoni dont know, was it theseven who brought it up last week when i started this?
11:23:16wodzJdGordon: he was talking about double buffering, doesn't he?
11:23:27JdGordonthat too
11:24:39wodzHonestly I can't imagine how this is going to help with dma. Lowlevel driver expects the data in framebuffer[] anyway
11:25:30JdGordonyes, that is one problem with all the drivers, they will need to be changed to deal with a pointer instead of that static array
11:25:49wodzthat is going to be tricky
11:26:28wodzsome framebuffers are located in iram. Iram isn't in buflib pool I guess
11:27:21JdGordonthat just means updates to the backdrop layer will be slower
11:27:22JdGordonthat layer though is never sent to the lcd
11:27:48wodzcoldfires (without dcache) will suffer badly if framebuffer is taken out of iram
11:27:56 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:28:43 Join factor [0] (~factor@
11:38:25 Quit ReimuHak_ (Read error: Connection reset by peer)
11:40:15 Join ReimuHak_ [0] (
11:52:51 Quit hilbert (Quit: Textual IRC Client:
11:53:43 Quit Xerion (Ping timeout: 255 seconds)
12:05:19 Quit GeekShadow (Ping timeout: 248 seconds)
12:06:09 Join Xerion [0] (
12:09:28JdGordonkugel: ping?
12:15:09dfkti made small images of the clip zip and sansa connect for the themes/manual site - who shoulod i contact to get them up?
12:16:02JdGordonthat would be Zagor or Bagder
12:16:53dfktthanks, both can consider themselves pinged, i guess :)
12:17:04 Join hilbert [0] (
12:17:43 Quit nick-p (Quit: Leaving)
12:17:52 Quit wodz (Quit: Leaving)
12:29:24 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
12:29:49kugelpJdGordon: pong
12:30:57JdGordonkugelp: what needs to be done to add transparency to an image? it looks like it requires the alpha data straight after the image data, so it cant be added after load for album arT?
12:34:58kugelpit sure can if you reserved/alloc enough space
12:35:38JdGordonwell, aa is stored on the audiobuf, not buflib directly
12:36:43kugelpthat doesn't matter
12:38:00JdGordonok, so i need to unconditioannly overalocate to hack this in?
12:38:03 Join TheLemonMan [0] (
12:38:09JdGordonwhat does the alpha buffer look like?
12:39:14kugelpits part of the bitmap data buffer, starting at alpha_offset. 4bit per pixel
12:39:47JdGordon0 is full transparent or full opaquw?
12:40:17 Join GeekShadow [0] (
12:42:26JdGordonok, so i just need to add an extra width*height/2 bytes on the end and memset that to somthing like 0x44 to get 50% transparency?
12:43:17dfktZagor, Bagder - i made pics for the fuze+ and sa9200 too
12:45:24dfktZagor, Bagder - here's the link to all four -
12:47:09JdGordonkugelp: do i need to modify anything else in the bmp struct or just the alpha_offset?
12:54:15JdGordonwhy the heck is there screens.bmp_part() and screens.bitmap_part()!!!!
12:59:21*JdGordon woners if he is completly misunderstanding the trasnparency thing
13:02:16JdGordonnope, just the usual stufup :)
13:03:06JdGordon <- 50% AA transparency
13:04:00 Join MethoS- [0] (~clemens@
13:06:50***Saving seen data "./dancer.seen"
13:15:13 Quit hilbert (Quit: Computer has gone to sleep.)
13:26:42kugelpJdGordon: see the big comment above the alpha stuff:
13:27:19kugelpJdGordon: that looks crap :)
13:30:23JdGordonhmm, any raosn bmp->alpha_offset is only checked for > 0?
13:31:50JdGordonapparently not
13:32:06kugelpwhat do you mean?
13:32:29JdGordonI've got the alpha data in a seperate handle which will always be < bmp->data
13:32:50kugelpalpha_offset isnt a pointer, that's not going to work is it?
13:33:07JdGordonexactly, its an offset, so there is no reaosn it cant be < 0
13:33:41kugelpthe reason is that it's supposed to be part of the data buffer
13:34:20JdGordonchange %Cd to %Cd(number) where number is the percentage you want
13:34:53kugelpseems pretty useless to me tbh
13:35:25JdGordonim not necessarily disagreeing
13:35:44kugelpanyway, for constant transparency an extra buffer is pretty wasteful. there should be a viewport field instead probably (like bg_pattern)
13:36:16JdGordonit isnt part of the viewport
13:36:31kugelpor of the bitmap struct
13:36:32JdGordonbut yes, 8K of constant is pretty crap
13:37:02 Join hilbert [0] (
13:41:18JdGordonlcd_bitmpa_part_mix() scares me
13:43:29*GodEater_ wonders about the 3.10 regression email
13:43:44GodEater_we haven't released RaaA at all on any platform have we?
13:43:48GodEater_so what's he on about?
13:45:17JdGordonis 0xaa (~66%) called 66% opaque or 66% transparent?
13:46:10 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
13:46:55kugelJdGordon: is the comment I directed you to unclear?
13:47:10JdGordonhanvt clicked it yet
13:47:33kugelthen do it before asking more questions about the internals please
13:47:45JdGordonYes SIR!
13:49:44TorneGodEater_: to be fair, that's not a thing a user has much reason to think about. they had an old version that worked better, and a new version that works worse
13:50:22Tornewhether it's "released" or "supported" is ultimately just an explanation for why the developers have chosen not to care about their issue
13:50:26GodEater_right - but I'm wondering where he got something called 3.9 from?
13:50:37GodEater_haha - true that ;)
13:50:48kugelTorne: he's not a user :)
13:50:54 Quit ReimuHak_ (Quit: Leaving...)
13:50:56kugelI mean he's also a dev
13:50:57Tornekugel: same thing, though
13:51:03Torneif you're asking someone else then you're a user :)
13:51:26TorneGodEater_: built it himself, or just means "trunk from before 3.10 branched"
13:51:41GodEater_yeah I guess so
13:51:46Tornethe latter is why i want to go for a better versioning system with git
13:51:54Tornesuch that trunk has a version number that makes sense to humans
13:51:57GodEater_talking of which? :)
13:52:06Tornedoing it!
13:52:08Tornesort of.
13:52:13Tornei poked at versioning yesterday for a bit
13:52:17Tornelunchtime, anyway :)
13:53:03gevaertsTorne, GodEater_: this isn't a user
13:53:16GodEater_I didn't think he was either
13:53:21GodEater_but I didn't know for sure
13:54:09 Join LinusN [0] (
13:54:46JdGordonAny objections to adding this AA transparency patch? its a bit of a ram waste, but only if you actually use it
13:54:52JdGordonsomeone can make it better later
13:56:25kugelwell, we've been adding new tags "because we can", why not params as well
13:56:31kugelI don't see it useful in any way
13:57:07kugelbut wouldnt it make more sense in %Cl, instead of %Cd?
13:59:40JdGordonnot necessarily... ideally we'd allow multiple %Cd tags in a skin
13:59:56JdGordonbut also its alot harder to add it to the load tag
14:00:15 Join teru [0] (
14:00:15kugelit is?
14:00:51kugelyou can have multiple %Cd, cant you?
14:00:59JdGordon%Cl() already has 2 optional params... and no you cant
14:01:04JdGordonwell, you can, but they wont all work
14:01:37kugelI think you can, and you cannot specifiy a different transparency for each
14:02:20JdGordontrust me, you cant
14:02:29JdGordonand that would be why I'd put it in the cd tag
14:03:04kugelyou cant have conditional viewports with %Cd in each?
14:03:24JdGordonyou can, but they wont work
14:03:29kugelwhy not?
14:03:54JdGordonbecause the aa struct needs the viewport it is to be drawn in
14:04:28JdGordonAA handling needs to be reworked anyway
14:06:15kugelI wonder why it is that way. I would have expected it to work
14:07:16JdGordonbecause images are drawn really stupidly.. which is what I've been bitching about for ages but can't change because it breaks all the cabbies
14:11:05JdGordonthis is for sure pointless without the layering stuff though
14:29:26 Quit bluebrother (Disconnected by services)
14:29:27 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
14:31:07 Quit hilbert (Quit: Computer has gone to sleep.)
14:32:23 Join hilbert [0] (
14:33:03 Quit fs-bluebot (Ping timeout: 248 seconds)
14:34:11 Join fs-bluebot [0] (
14:47:22 Join ReimuHak_ [0] (~reimu@
14:47:23kugelJdGordon: cabbie can be fixed
14:47:31kugelthat's a poor excuse :P
15:06:52***Saving seen data "./dancer.seen"
15:30:56 Join y4n [0] (y4n@unaffiliated/y4ndexx)
15:44:01 Join Thra11 [0] (~thrall@
15:55:44 Quit teru (Quit: Quit)
16:05:10 Join WalkGood [0] (~4@unaffiliated/walkgood)
16:10:18 Quit ReimuHak_ (Quit: Leaving...)
16:15:33 Quit TheLemonMan (Quit: Ex-Chat)
16:16:44 Join casainho [0] (
16:25:02 Join jlbiasini [0] (
16:29:30 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
16:29:43 Join ChickeNE_ [0] (~ChickeNES@
16:32:22funmanwodz: pong
16:43:12 Join gbl08ma [0] (~gbl08ma@
16:49:23 Quit hilbert (Remote host closed the connection)
16:50:52 Join antil33t| [0] (~antil33t@
16:52:34 Quit antil33t (Ping timeout: 252 seconds)
16:56:46 Part Zagor
16:57:53 Join antil33t [0] (~antil33t@
16:59:39 Quit antil33t| (Ping timeout: 260 seconds)
17:06:56***Saving seen data "./dancer.seen"
17:07:58 Quit antil33t (Ping timeout: 252 seconds)
17:09:42 Join antil33t [0] (~antil33t@
17:19:30 Part LinusN
17:22:50 Join kadoban [0] (
17:23:38 Quit ChickeNE_ (Quit: Computer has gone to sleep.)
17:45:58 Join ChickeNE_ [0] (~ChickeNES@
17:48:52dfktwhat could be the cause for audio balance shifting slightly on sansa ams players when the volume is changed? i.e, cause of a broken ear, i have balance on -9 to the left, and when i change the volume one notch, it shifts a bit to the right, one notch more, and it shifts back to the left
17:48:58 Join hilbert [0] (
17:49:33dfktthis happens at various volume levels
17:49:56gevaertsSounds like a rounding issue
17:50:10dfktah, iirc the ams chips can only do 1.5db steps?
17:51:38 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:52:58 Quit ej0rge (Ping timeout: 245 seconds)
17:54:11 Join ej0rge [0] (~alhaz@
17:56:31 Quit kadoban (Remote host closed the connection)
17:59:35 Join liar [0] (
18:18:04funmanbalance is in 1 steps instead of 1.5 ?
18:27:16 Join wodz [0] (
18:27:47 Quit antil33t (Ping timeout: 240 seconds)
18:27:53wodzfunman: do you know of some guide how to setup and use eda other than geohot's youtube video?
18:28:47funmanyou have to run eda and point your browser to the port it opens
18:31:06wodzwhat kind of files does it support? elf, plain bin or something preprocessed?
18:32:07funmanprobably bin
18:33:11 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:33:42 Join antil33t [0] (~antil33t@
18:37:30funmanwow no Makefile
18:37:41funmanwodz: which version of eda do you build?
18:40:06wodzthe one from the git repo you provided yesterday
18:40:13wodzfunman: this is cmake project
18:41:40funmanthe HEAD doesnt build
18:41:59 Join GodEater [0] (
18:41:59 Quit GodEater (Changing host)
18:41:59 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
18:43:05wodzfunman: no surprise for project from 2009
18:43:38funmannope it s just that geohot always forget to do git add ...
18:46:42wodzI am kinda interested but today is not good day for me so I leave this for a moment
18:48:40wodzfunman: do you know of particular version/revision which actually works?
18:49:13funmanwodz: i'm trying to fix HEAD~2
18:50:46wodzcan't make linux bch library to produce the same ecc bytes as rk sdk :/
18:51:11funmanwodz: git pull again and check branch fixup
18:54:30wodzahh it has input file hardcoded in
18:54:54wodzas well as memmap it appears
18:56:00funmansegfualts for me :/
19:05:45wodzthere seems to be some earlier project eda-reverse also
19:06:19funmanyep, eda-2 was rewrite of eda-reverse
19:07:00***Saving seen data "./dancer.seen"
19:07:25 Quit antil33t (Ping timeout: 244 seconds)
19:09:32funman==18215== total heap usage: 528,402 allocs, 55 frees, 186,864,900 bytes allocated
19:10:32wodzslightly unbalanced mallocs and frees :-)
19:14:25 Join Strife89 [0] (~Strife89@
19:15:31 Join lebellium [0] (
19:17:26 Quit ChickeNE_ (Quit: Computer has gone to sleep.)
19:19:43 Quit dreamlayers (Quit: Bye)
19:31:17 Join antil33t [0] (~antil33t@
19:32:34 Join ReimuHak_ [0] (~reimu@
19:36:11 Quit ps-auxw (Ping timeout: 240 seconds)
19:37:24 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
19:38:25 Quit Strife89 (Quit: Every time you make a typo, the errorists win.)
19:38:38 Join Strife89 [0] (~Strife89@
19:39:24CIA-10New commit by pamaury (r31098): fuze+: update keymap and touchpad driver based on FS #12405 with minor modifications. Thanks to Jean-Louis Biasini.
19:41:18CIA-10r31098 build result: 2 errors, 0 warnings (pamaury committed)
19:42:08 Quit antil33t (Ping timeout: 252 seconds)
19:44:48 Quit sinthetek (Ping timeout: 258 seconds)
19:45:00CIA-10New commit by pamaury (r31099): mkimxboot: remove an unecessary include that will prevent compilation of rbutil
19:45:02 Quit ReimuHak_ (Ping timeout: 245 seconds)
19:46:39CIA-10r31099 build result: 2 errors, 0 warnings (pamaury committed)
19:47:32CIA-10New commit by pamaury (r31100): imx233: add power debug screen
19:49:12CIA-10r31100 build result: 2 errors, 0 warnings (pamaury committed)
19:54:39 Join TheLemonMan [0] (
19:55:29gbl08mathanks to the new implementation of some of the math lib functions in LUA for Rockbox, I could now port aeslua to Rockbox with some little modifications
19:55:54 Join antil33t [0] (~antil33t@
19:56:25gbl08mabut there's a problem: it looks like when a file is encrypted, it can't be decrypted. perhaps the aes encryption is not valid due to things like math.ceil being used in a fixed-point system
19:56:47gbl08maor else, it's because I'm not writing/reading to the files in binary mode
19:57:50gevaertsI assume we're not allowed to ask questions like "Why do AES in an interpreted language on a slow DAP?"?
20:00:37 Quit antil33t (Ping timeout: 255 seconds)
20:02:10 Quit lebellium (Remote host closed the connection)
20:02:45 Join lebellium [0] (
20:04:07 Join bertrik [0] (
20:04:07 Quit bertrik (Changing host)
20:04:07 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
20:08:26gbl08mano you're not. :) but I'll answer: I'm making this for fun, and I wonder how fast it will be. So far, the script could encode rb.txt in less than 10 seconds, which is good :)
20:08:39 Join ReimuHak_ [0] (~reimu@
20:10:33 Join Horscht [0] (
20:10:36 Quit Horscht (Changing host)
20:10:36 Join Horscht [0] (~Horscht@xbmc/user/horscht)
20:11:36gbl08maactually, the script takes more time loading the LUA scripts, than executing them to encrypt the file.
20:16:22gbl08maI made it! I got wrong some variable names, after fixing them, I can now encrypt a 2632 bytes file to AES-128 in less than 5 seconds, and decrypt it back in less than 5 seconds too.
20:17:07gbl08mano data loss, and to the naked eye, the decoded file has the very same content as the original (goes checking with md5 plugin...)
20:18:38gbl08manope... the md5sum is different :(
20:20:19gbl08mahmmm... the decoded file misses 6 letters and one line break, at the end of the file. so, it misses 7 bytes, right?
20:27:42 Join antil33t [0] (~antil33t@
20:28:40gbl08maI encoded a BMP file and then decoded it. although Rockbox couldn't read the decoded picture (perhaps because some bytes are missing at the end), the image viewer of Gnome can read it without problems.
20:30:19funmanperhaps gnome is less strict than us, wouldn't surprise me
20:30:33gbl08malooks like the byte loss increases with the file size, the image is 68KB and it lost 14 bytes after having been encoded and decoded
20:30:39dfkt<funman> balance is in 1 steps instead of 1.5 ? <−− not every step on the ams sansas actually change balance, yes. sometimes it's two steps, sometimes it's one - kind of interpolated. markun once started a patch to fix that, but that was maybe a year ago:
20:30:54gevaertsgbl08ma: I'd guess it's to do with partial last blocks
20:31:04gevaertsNot with actual filesize as such
20:32:11 Join sinthetek [0] (
20:32:11 Quit sinthetek (Changing host)
20:32:11 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
20:32:30 Join kadoban [0] (
20:32:32gbl08mabtw, it took around two minutes to encode or decode the file. obviously, this script would never be used for encrypting sound files
20:33:14gbl08ma(encrypting sound could be useful e.g. on targets with recorder... nobody would be able to read the records without the password)
20:33:58gevaertsYes, but you'd want realtime encrypting then
20:35:24gbl08mawell, people can record and then encrypt. of course, the "secret" would be exposed while you don't encrypt it.
20:37:20gevaertsIt will be exposed until you securely overwrite the disk location
20:38:06gevaertsIf you want this for serious reasons, you want to encrypt before you hit the disk at all
20:38:24gbl08maso, the next step is to write a (LUA? :) ) plugin that fills the empty disk space with zeros :)
20:39:31gevaertsExcept you can't, really, 100%. There are things like remapped bad sectors, the unused bits at the end of FAT clusters, and FTLs
20:39:42gbl08mamy idea for using AES is to keep safe things like the keybox data file. TEA, which is what keybox uses, is not that hard to crack
20:39:59 Part WalkGood
20:40:30gbl08maif you are so worried about security that you can't withstand a few (bad) sectors left without being overwritten, you shouldn't use computers in the first place :)
20:40:54gevaertsMaybe :)
20:41:32gevaertsBut if doing things properly isn't too much extra work...
20:42:11gbl08maI'm not sure, but I think the "forgetting bytes" problem is not related with aeslua. perhaps it has something to do with floating point emulation in a fixed point system.
20:42:45funmangbl08ma: are you using floats?
20:43:38gbl08malua uses doubles. lua's bitlib needs to convert everything to a double for providing bitwise operations
20:44:07gbl08maI don't know what doubles are converted to in Rockbox's C lua parser
20:44:50wodzekhm, bitwise operations on doubles?
20:44:59gbl08manor I do know how things like math.ceil are done in Rockbox (they only got implemented in rockbox recently, who committed?)
20:46:07funmanstatic int math_ceil (lua_State *L) { /* Doesn't change anything in fixed point arithmetic */
20:46:22funmani think we simply don't implement float or doubles
20:46:43funmanthe mathlib is actually 2 years old and wasn't touched since initial commit of lua
20:47:40gbl08mabut I don't think it supported all these functions last time I played with LUA on Rockbox: abs ceil deg floor fmod max min rad random randomseed
20:48:03 Quit antil33t (Ping timeout: 258 seconds)
20:48:18gbl08maIIRC, it only supported "random" and "randomseed"
20:48:33funmanceil doesn't make sense if you use integers
20:48:51jlbiasiniis there any way to generate the keymaps file for manuals from the one in rockbox? I just spend a whole week-end on mapping all the fuze+ keys noway that I do that again for manuals keymaps. I would even prefen to spend a week on writing an aplication for that!
20:49:02funmanbut it was supported still. rounding of an integer is the integer itself
20:50:13 Join antil33t [0] (~antil33t@
20:51:16gbl08mawell, in lua if i use numbers that are not integers, it still works, except that the result is wrong
20:51:29gbl08maI once tried to convert celsius to fahrenheit...
20:51:49gbl08mait couldn't divide 5 by 2 properly, of course
20:52:07funmanjlbiasini: have you looked at model specific files for the manual?
20:52:58jlbiasiniyeah there are thos do you mean those?
20:59:11jlbiasinithat's the reason I'm asking do I have to write those or are they automaticly generated if I configure something as it should be ?
21:00:07funmanyou have to write those but writing an automatic generator would be better of course
21:00:38jlbiasinithen I'll do it
21:00:45funman!if you're not sure you can check if the file is versioned, if it is it is not generated
21:01:03funmangit log keymap-sansafuze.tex shows 6 revisions
21:01:51 Quit casainho (Ping timeout: 258 seconds)
21:02:33jlbiasiniI think it must be automaticly generated because I red somewhere that it was the point of have versionned manuals that the keymap get updates
21:02:57jlbiasininow that I remenber it
21:03:43funmanjlbiasini: no it's not, it needs manual modifications
21:03:54funmanand creation of course
21:05:18 Quit antil33t (Ping timeout: 258 seconds)
21:06:09 Join antil33t [0] (~antil33t@
21:06:14jlbiasiniso I will investigate on something to have them automaticly generated it might takes mouth
21:06:24 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
21:06:40 Join robin0800 [0] (
21:06:49 Quit y4n (Quit: 6,000,000 ways to die — choose one.)
21:06:52jlbiasinibut the fuze+ is not that advanced anyway we are still missing an installation method for rbutil
21:07:03***Saving seen data "./dancer.seen"
21:07:21jlbiasiniok new task for me \o/
21:14:07 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
21:15:49 Quit antil33t (Ping timeout: 244 seconds)
21:17:34 Part jlbiasini
21:21:09 Quit robin0800 (Ping timeout: 244 seconds)
21:26:23 Quit ReimuHak_ (Quit: Leaving...)
21:28:53 Join antil33t [0] (~antil33t@
21:32:23 Join unlink_ [0] (6d5b6ac7@gateway/web/freenode/ip.
21:32:24 Join jlbiasini [0] (
21:33:59 Part jlbiasini
21:36:09 Quit antil33t (Ping timeout: 260 seconds)
21:37:42gevaertsunlink_: that looks interesting...
21:37:53*gevaerts pings domonoky and bluebrother^ for this
21:40:10 Quit liar (Quit: hallowed are the ori!)
21:40:23 Join liar [0] (
21:42:10 Join robin0800 [0] (~robin0800@
21:43:50saratoga_does it happen with the current build?
21:49:56*bluebrother^ appears
21:50:17bluebrother^unlink_: that's on Linux, right?
21:50:45bluebrother^do other statically linked
21:50:49bluebrother^Qt programs work?
21:51:24bluebrother^I'm under the impression that this is caused by some (missing?) Qt configuration and thus it using the wrong fonts
21:51:34bluebrother^unlink_: well, _statically_ linked programs?
21:52:35bluebrother^that doesn't count. It was built on a different setup ;-)
21:53:52bluebrother^hmm. Try running qtconfig-qt4 (or whatever that is named on your system) and check for the font settings.
21:54:28 Quit Thra11 (Ping timeout: 255 seconds)
21:56:14 Quit liar (Ping timeout: 248 seconds)
21:56:57 Join antil33t [0] (~antil33t@
21:58:29bluebrother^I need to check the differences between the two build setup then.
21:58:44bluebrother^in the meantime, since you're on Gentoo building it youself should be the simplest solution :)
22:01:45 Quit robin0800 (Read error: Connection timed out)
22:02:02gevaertsWasn't there a theme engine change a while ago that had some impact on progress bars?
22:02:14bluebrother^hmm. Maybe it's fontconfig related like the problem reported here:
22:03:46*bluebrother^ fires up some VWs
22:05:51bluebrother^it could indeed be fontconfig related. The old build machine has fontconfig development headers installed while the new one doesn't
22:06:28 Join Thra11 [0] (~thrall@
22:06:46bluebrother^the bad thing is that trying that out means I need to rebuild Qt which takes a while
22:07:38bluebrother^well, building Qt usually takes a few hours
22:08:20bluebrother^unlink_: I can build a new binary by tomorrow evening. Interested in showing up here and giving it a try?
22:10:29bluebrother^broken in what way?
22:10:41bluebrother^ah, found it in the backlog :)
22:11:03bluebrother^does music actually play?
22:12:01 Join Keripo [0] (
22:12:05bluebrother^I don't have a real idea on this, I'm not really involved in Rockbox UI code.
22:13:28bluebrother^JdGordon is working on the skin stuff, you can try to catch him
22:13:58 Quit froggyman (Quit: People who think they know everything are really annoying to those of us who actually do.)
22:14:28 Join froggyman [0] (~froggyman@unaffiliated/froggyman)
22:21:09 Quit antil33t (Ping timeout: 244 seconds)
22:23:02 Quit TheLemonMan (Quit: Ex-Chat)
22:32:32 Quit unlink_ (Ping timeout: 265 seconds)
22:36:59 Quit jhMikeS (Read error: Connection reset by peer)
22:38:32 Join antil33t [0] (~antil33t@
22:59:53 Join jlbiasini [0] (
23:07:04***Saving seen data "./dancer.seen"
23:07:33 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
23:09:53 Quit wodz (Read error: Operation timed out)
23:16:51 Quit MethoS- (Remote host closed the connection)
23:20:29 Quit Horscht (Quit: Verlassend)
23:28:49 Join Topy44 [0] (
23:29:02 Quit hilbert (Quit: Textual IRC Client:
23:29:02 Quit T44 (Read error: Connection reset by peer)
23:33:25 Join Scromple [0] (
23:34:16 Quit pamaury (Remote host closed the connection)
23:36:40 Quit Keripo (Quit: Leaving.)
23:37:02 Join ReimuHak_ [0] (
23:43:12 Quit tjb0607 (Read error: Connection reset by peer)
23:43:37 Join nosa [0] (
23:46:44 Quit factor (Read error: Connection reset by peer)
23:46:55 Quit nosa-j (Ping timeout: 244 seconds)
23:46:56 Nick nosa is now known as nosa-j (
23:49:52 Join factor [0] (~factor@
23:57:05 Join antil33t| [0] (~antil33t@
23:57:50 Quit antil33t (Ping timeout: 260 seconds)

Previous day | Next day