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

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

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

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2012-03-19

00:07:49 Quit prof_wolfff (Ping timeout: 246 seconds)
00:12:57 Quit enthdegree (Ping timeout: 244 seconds)
00:18:22 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
00:20:01 Quit n1s (Quit: Ex-Chat)
00:21:14 Quit Scromple (Quit: Leaving)
00:23:09 Join prof_wolfff [0] (
00:27:25 Quit prof_wolfff (Ping timeout: 246 seconds)
00:29:02 Quit liar (Remote host closed the connection)
00:37:59 Quit enthdegree (Ping timeout: 265 seconds)
00:41:17***Saving seen data "./dancer.seen"
00:43:06KiwiCam[Saint]: I'm trying to break your code down. I have these %xl's defined. All single image bitmaps no larger than 10 wide x 7 high, repeat,shuffle,charge,charging. This works: %Vl(a,117,43,11,8,-) %?bp<%?bc<%xd(charging,1)|%xd(charged,1)>> If I then start adding the %and's such as: %?bp<%?bc<%?and(%mm, %ps)<%t%xd(charging,1);%t%xd(shuffle,1);%t%xd(repeat,1)|%?mm<%xd(shuffle,1)>|%xd(charged,1)>> I come unstuck. Of course it's the < &> signs, but I can't f
00:43:06KiwiCamigure it out. I've tried your full line of code, but that's not working (%pr s/b %ps BTW)
00:47:40 Join prof_wolfff [0] (
00:49:00 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
00:49:58[Saint]Its fairly easy, instead of looking at it as one huge line, look at the seperate cases you want to handle, one at a time. Then encapsulate it in %?bp<%?bc<stuff you want to do while sharging|stuff to do when charged>|stuff to do at all other times>
00:50:58[Saint]when you have working lines to handle each of those cases, then elclose it in the appropriate condition.
00:51:52[Saint]If for some reason you can't nest %?and statements inside an arbitrary condition (which I haven't tried, but can't immediately see failing), I'd be willing to call that a bug.
00:53:06KiwiCamOK. I'm happy. I haven't tried the %and on it's own. So, I'll do that.
00:53:53 Quit lebellium (Quit: ChatZilla [Firefox 12.0/20120314195616])
00:55:16 Quit bertrik (Quit: And That, My Liege, Is How We Know the Earth to Be Banana Shaped)
01:01:44 Join bitcraft [0] (
01:03:06 Part Zagor
01:26:56 Quit prof_wolfff (Ping timeout: 276 seconds)
01:40:50 Quit evilnick (Quit: Bye)
02:02:42 Quit perrikwp (Ping timeout: 252 seconds)
02:06:49 Quit soap (Ping timeout: 246 seconds)
02:07:16 Quit MethoS- (Ping timeout: 272 seconds)
02:13:21 Join anewuser_ [0] (~anewuser@
02:17:12 Quit anewuser (Ping timeout: 245 seconds)
02:25:32 Quit funman (Ping timeout: 246 seconds)
02:27:27 Join funman [0] (~fun@rockbox/developer/funman)
02:28:06 Join perrikwp [0] (
02:41:21***Saving seen data "./dancer.seen"
03:35:53 Quit enthdegree (Ping timeout: 260 seconds)
03:36:10 Quit bitcraft (Remote host closed the connection)
03:38:32 Join soap [0] (
03:38:32 Quit soap (Changing host)
03:38:32 Join soap [0] (~soap@rockbox/staff/soap)
03:54:13 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
04:05:36 Quit amiconn (Disconnected by services)
04:05:36 Quit pixelma (Disconnected by services)
04:05:37 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:05:39 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:05:41 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:05:58 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:24:01 Quit [7] (Disconnected by services)
04:24:11 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:25:46 Join Scromple [0] (~Simon@
04:41:23***Saving seen data "./dancer.seen"
05:02:40enthdegreeThe MEElec page is getting speedydeleted
05:31:11 Quit anewuser_ ()
05:31:37 Quit TheSeven (Ping timeout: 240 seconds)
05:38:44 Quit curtism (Quit: Live Long and Prosper)
05:57:05 Join Rob2222 [0] (
06:00:27 Quit Rob2223 (Ping timeout: 245 seconds)
06:04:37 Quit mc2739 (Ping timeout: 240 seconds)
06:06:30 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
06:13:11 Quit Rower85 (Read error: Connection reset by peer)
06:20:41 Quit factor (Ping timeout: 265 seconds)
06:28:47 Join msh [0] (msh@
06:41:25***Saving seen data "./dancer.seen"
07:01:20 Quit KiwiCam (Quit: Leaving)
07:04:18 Quit enthdegree (Ping timeout: 264 seconds)
07:13:04 Join factor [0] (
07:15:54 Join leavittx [0] (~leavittx@
07:16:07 Join KiwiCam [0] (~KiwiCAM@
07:19:03 Quit msh (Disconnected by services)
08:14:53 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:27:40 Join LinusN [0] (
08:33:31 Join einhirn [0] (
08:40:19 Join wodz [0] (
08:41:26***Saving seen data "./dancer.seen"
08:42:05 Join bertrik [0] (
08:42:05 Quit bertrik (Changing host)
08:42:05 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:46:52CIA-44Commit e1796fb in rockbox by 03Marcin Bukat: mrobe500: fix warrnig in bootloader
08:46:53CIA-44Commit 395290b in rockbox by 03Marcin Bukat: fix cache function name
08:46:55CIA-44Commit 69978d7 in rockbox by 03Marcin Bukat: center logo horizontaly in bootloader version of show_logo()
08:47:11 Join ender` [0] (
08:50:02CIA-4469978d7 build result: All green
08:53:03wodzRegarding FS #10776, isn't it enough to call commit_discard_idcache() *before* copying rb image from audiobuf to the beginning of DRAM?
08:53:03fs-bluebot 3ROLO issues on ARM v4/v5 targets (bugs, new)
09:03:11 Join mortalis [0] (~mortalis@
09:12:01*wodz is kindly asking for review of g#190
09:18:55Zagorgevaerts: can you benchmark the ypr0 build? I have a feeling it's rated too low.
09:25:28 Quit bertrik (Read error: Operation timed out)
09:25:42wodzif I want to read multiple words from cons address (aka do not change pointer) on ARM ldmia rX, {....} should do right?
09:26:17 Join jlbiasini [0] (
09:26:32 Part jlbiasini
09:39:20 Join nick_p [0] (
09:44:52 Join megal0maniac [0] (
09:47:46 Quit wodz (Quit: Leaving)
09:48:47 Join jlbiasini [0] (
10:00:59 Quit kadoban_ (Ping timeout: 252 seconds)
10:02:02 Quit megal0maniac (Read error: Connection reset by peer)
10:02:09 Join megal0maniac_ [0] (
10:04:16 Quit megal0maniac_ (Client Quit)
10:04:33 Join megal0maniac [0] (
10:04:33 Join diginet [0] (
10:08:22 Join megal0maniac_ [0] (
10:08:22 Quit megal0maniac (Read error: Connection reset by peer)
10:09:57 Join MethoS- [0] (~clemens@
10:10:56 Quit megal0maniac_ (Read error: Connection reset by peer)
10:12:22 Join megal0maniac_ [0] (
10:32:00 Join wodz [0] (
10:41:28***Saving seen data "./dancer.seen"
10:44:59KiwiCam[Saint]: Update. The %and is what seems to be the issue. I can get this to work for instance, or at least, it displays the icon: %?and(%mm, %ps)<%xd(shuffle,1)> This, however, will not: %?and(%mm, %ps)<%t%xd(shuffle,1);%t%xd(repeat,1)>
10:49:22 Quit megal0maniac_ (Quit: Konversation terminated!)
10:54:07JdGordonthe %t is redundant
10:54:11JdGordonand why not?
10:54:43JdGordonerr.. duh... %t like that is illegal
10:54:56JdGordonuse −−debugwps in the sim to see the parser errors
10:55:09JdGordonKiwiCam: ^
10:55:09[Saint]why can't you do nested conditional sublines?
10:55:23[Saint]I'd call that a bug if its "illegal".
10:55:42JdGordon%t requires a number
10:55:54[Saint]No it doesn't, or, when did you change it?
10:56:00[Saint]It should default to 1 second.
10:56:27JdGordon{ SKIN_TOKEN_SUBLINE_TIMEOUT, "t" , "D", 0 },
10:56:31JdGordonit requires a number
10:56:40[Saint]It never used to.
10:57:17[Saint]the same with all timeout style tags, if you neglect the timeout, it should default.
10:57:22JdGordonit changed with the great skin breaking change commit of whenever it was
10:57:23[Saint]consistency, yo.
10:57:24KiwiCamI'd like to swap between the icons, rather than display them side by side.
10:59:35KiwiCamAs an aside. Where am I going wrong here: %?cf<%xd(AlienClock,%cH)|%xd(AlienClock,%cI> or with %ci Anything obvious? This works: %?cf<%xd(AlienClock,%cH)|>
11:00:31[Saint]what....why are you passing a tag to the subimage field?
11:00:47JdGordonKiwiCam: use the simulatror with −−debugwps becasue you have syntax errors
11:01:38[Saint]Ohhh...I see now.
11:02:06KiwiCamJdGordon: I'm not getting the debug window appearing. I think it's a compiz issue.
11:02:28KiwiCam[Saint]: Tag?
11:02:39[Saint]Don't worry.
11:03:07JdGordonKiwiCam: arg, you get the output in a console (so dont start it from a shortcut)
11:03:11*JdGordon gone
11:05:02KiwiCamJdGordon: I'm not. It's from the .exe I'll have a look into why the console isn't appearing...
11:05:26 Join petur [0] (~petur@rockbox/developer/petur)
11:06:28nick_pAs requested, I've made g#182 an option. Assuming there's no last minute reviews, I plan to commit it after a bit more testing.
11:06:29fs-bluebot3Gerrit review #182 at,182 : Option to constrain get_next_dir() to directories below global_settings.start_directory. by Nick Peskett (changes/82/182/5)
11:06:39[Saint]%?some_tag<%xd(some_identifier,some_tag)|%xd(some_identifier,some_tag)> is perfectly valid.
11:06:56[Saint]but in your example, you're missing a closing bracket.
11:08:51KiwiCamFaaarge. [Saint] Thanks. So, now it's just the %and issue.
11:10:07KiwiCamand the console. It's not compiz. I had toldcompiz to give my WINE windows background focus, but the console just doesn't appear. It does appear on my laptop though.
11:10:34 Join megal0maniac [0] (
11:11:57 Quit jhMikeS (Ping timeout: 276 seconds)
11:12:11*gevaerts wouldn't actually recommend using windows builds on linux
11:12:18[Saint]nor I.
11:12:29[Saint]I just clicked how horribly ugly that is.
11:12:51[Saint]Build your own from source, it'll take like half an hour.
11:12:56KiwiCamSo, where do I get the Linux build?
11:13:08[Saint]You build it.
11:13:37gevaertsHalf an hour?
11:13:44*gevaerts tells [Saint] to get a faster CPU
11:13:55[Saint](+getting sources)
11:13:58KiwiCamWhere's the code?
11:13:59gevaertsAh, ok :)
11:14:10[Saint]{and accomodating for shit connection}
11:14:41[Saint]KiwiCam: seriously, now you're not even trying ;)
11:14:47[Saint]see: UsingGit in the wiki
11:15:14 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
11:15:53gevaertsNote that for just building you can probably stop reading quite soon. Most of that document is about how to get your changes back to us, just getting the source is quite simple
11:17:01 Join jdgord [0] (~jdgord@rockbox/developer/JdGordon)
11:17:29*[Saint] assumed that was obvious, but, good call gevaerts ;)
11:17:53[Saint]Better speak up, lest we get a sim build submitted to gerrit :P
11:18:04KiwiCamI figured that the Development stuff was just for developing :(
11:18:22[Saint]What do you think building a sim is?
11:18:24gevaerts[Saint]: I suspect it's obvious when you read it. I just want to avoid people getting discouraged by the length of the page :)
11:19:27KiwiCamI always wondered where the Linux version of the RB sim was. The Windows one has always work for me though, mostly.
11:20:17[Saint]I think its pretty much assumed that if you're running Linux, you're perfectly capable of building your own sim.
11:20:35[Saint]Its non-trivial for Windows, however.
11:20:44gevaertsPeople are free to provide linux builds if they want to. After all that's exactly what happens for windows builds
11:21:24KiwiCam[Saint]: I sort of figured that the "daily" Sim build were sort of pre-prepared :) I am clearly mistaken :)
11:22:15*[Saint] isn't so sure KiwiCam is aware of what he's saying before he hits enter.
11:22:24KiwiCam[Saint]: Linux can be for GUI users too :)
11:22:49[Saint]What the flying fudge are you talking about?
11:23:01[Saint]You're really not making any sense now.
11:23:34KiwiCam[Saint]: The SIM. The Windows version is always available. That's what I've been using.
11:24:01*[Saint] gives up.
11:24:11gevaertsWhat does that have to do with GUI users?
11:24:36gevaertsThe point is that building a sim on linux takes at most four commands, while on windows it's a weekend project
11:24:39 Join swilde [0] (
11:24:59gevaertsAnd even then, you might have noticed that rockbox does *not* provide windows sim builds
11:25:14[Saint]Apparently not.
11:26:22KiwiCamSo, what are these?">
11:26:41gevaertsThey are windows sim builds
11:27:42gevaertsWhat's your point there?
11:27:46[Saint]I think he's missing the point (and the header) that they're not Rockbox sanctioned.
11:28:01KiwiCamSo that's great. I have to learn how to use Git (that's something new) and then I presume I just do a make/install with the latest source code and I'm away.
11:28:45[Saint]You don't have to learn to use git at all.
11:28:57[Saint]You can copy/paste one command, that's it.
11:29:54KiwiCamI always figured that they were the source of the SIM. Clearly I was mistaken. I've never found the Rockbox site/Wiki to be very clear.
11:31:17[Saint]The sim builds are built from current source, but should be treated as unsupported builds.
11:31:53[Saint]Its also worth noting that they are not housed on the wiki at all.
11:32:23gevaertsThe *only* place on our wiki that links to rasher's sim builds says "However, unofficial (but trusted) pre-compiled Windows versions of the simulators can be found here". Can you suggest wording that makes it clear that they are unofficial (but trusted) builds?
11:32:41KiwiCamSo...Rockbox main page/ Wiki / Documentation Index / Using Git. It's all so clear to me now. I don't know how I missed that.
11:34:49gevaertsI agree, especially since that page is linked to from the UiSimulator page very near the top
11:36:01KiwiCamI'll have a go at compliing the code tomorrow. I'm too tired now. Thanks everyone. Honestly. That Sim pages is the only way I've been able to get any theming done, until now.
11:36:28 Join prof_wolfff [0] (
11:37:20KiwiCamAt least Rashers Sims work though :) Hours of fun :) So...
11:38:19 Part jlbiasini
11:38:21KiwiCamif that's the case then, I should be able to compile the code on my android and run the lastest code off of that, shouldn't I?
11:38:54KiwiCamRashers Sims - I should say mostly.
11:39:53jdgordYou're on Linux and not building the Sims manually?
11:39:57 Join lebellium_gs2 [0] (
11:40:45[Saint]And, if you're on Linux, why the *fudge* would you want to build in a chroot on Android?
11:40:52KiwiCamjdgord: No. Never new where or how to do this until now.
11:41:11[Saint](because, that's the only sure as shit won't build on Android natively)
11:41:47KiwiCam[Saint]: I just want to make some themes, while I have spare time at work. Clip themes, not Android ones.
11:42:19KiwiCam[Saint]: Oh yeah. Java.
11:45:24lebellium_gs2Hello JdGordon, would it be possible to add a 'random' tag to the skin engine? I'd like to display subimages at random
11:48:58lebellium_gs2Because it's too difficult to implement that?
11:49:42KiwiCamShouldn't I at least be able to get the %and working (as per [Saint]'s code on the actual Clip using the current build via the Rockbox Utility?
11:49:57CIA-44Commit be10817 in rockbox by 03Nick Peskett: Option to constrain get_next_dir() to directories below global_settings.start_directory.
11:53:02CIA-44be10817 build result: All green
11:53:02[Saint]You can easily implement your own psuedo-random condition.
11:53:43[Saint](using a substring of time elapsed/remaining, perhaps?)
11:55:49lebellium_gs2Hum don't know if that makes sense in my case. Actually I have several different designs for the no-album art pic and for the moment I display a different pic according to the file format (mp3, wma, flac etc).
11:56:29lebellium_gs2I'd like it not to depend on the file format but really random
11:58:01gevaertsMake them depend on the last digit of the track time or something like that
11:58:25gevaertsThat's still not random in the sense that it's always the same for a specific track, but still not bad I think
11:59:04lebellium_gs2Would already be better indeed
12:00:14lebellium_gs2I'll try that tonight and let you know
12:04:44lebellium_gs2Thanks guys and have a nice day
12:04:59 Quit lebellium_gs2 (Quit: Bye)
12:11:08 Quit leavittx (Ping timeout: 244 seconds)
12:11:52 Join TomColler [0] (
12:16:28wodzmortalis: ping
12:17:43 Join anewuser [0] (~anewuser@
12:17:51 Quit anewuser (Changing host)
12:17:51 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
12:35:02mortaliswodz: pong
12:41:31***Saving seen data "./dancer.seen"
12:44:59wodzmortalis: what is CPU_ID of your hifis? It should be printed HW info debug screen
12:50:34mortaliswodz: 0xA100027B
13:03:19 Quit nick_p (Quit: Leaving)
13:08:41 Quit megal0maniac (Quit: Konversation terminated!)
13:13:21 Join leavittx [0] (~leavittx@
13:13:26 Join jlbiasini [0] (
13:13:35 Part jlbiasini
13:20:31 Quit leavittx (Ping timeout: 272 seconds)
13:43:29 Join jlbiasini [0] (
13:44:37 Part jlbiasini
13:50:45 Quit jdgord (Quit: Bye)
13:52:57 Quit petur (Ping timeout: 245 seconds)
14:01:47 Join jlbiasini [0] (
14:02:01 Part jlbiasini
14:09:08 Join dfkt [0] (~dfkt@unaffiliated/dfkt)
14:15:54 Join Thra11 [0] (~thrall@
14:21:32 Join nick_p [0] (
14:41:32***Saving seen data "./dancer.seen"
14:47:54 Quit XavierGr (Ping timeout: 246 seconds)
14:54:55 Join WalkGood [0] (~4@unaffiliated/walkgood)
15:04:10 Quit TBCOOL (Quit: maintenance.)
15:16:30 Quit wodz (Quit: Leaving)
15:20:32 Join bitcraft [0] (~bitcraft@
15:45:47 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
15:48:17 Join Skimsi [0] (d4ba2c76@gateway/web/freenode/ip.
15:53:00 Quit Thra11 (Remote host closed the connection)
15:57:18 Join Rower85 [0] (
15:57:36SkimsiHi guys. I need some help with my Cowon D2. My screen is dead and now I want to make physical buttons work with this patch: As I have no working screen could someone take a quick look through the patch and tell me what I have to enter into config.cfg to activate the hold-buttons?
16:03:40 Join Thra11 [0] (~thrall@
16:05:57[Saint]"buttons in hold mode: on" looks like it should do it.
16:06:42[Saint]Skimsi: ^
16:15:28 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
16:17:16 Join TBCOOL [0] (
16:28:27SkimsiWorking! Thank you. Now at least it is useable. I just dont want to loose rockbox as my firmware and the only comparable alternative to a used D2 would have been a Cowon Z2 or something like that in terms of soundquality. I am so happy I dont have to go this way.
16:41:35***Saving seen data "./dancer.seen"
16:42:21[Saint]Happy to help Skimsi
16:45:15CIA-44Commit 9a03580 in www by 03Frank Gevaerts: Recalibrate all builds
16:55:50 Quit Skimsi (Ping timeout: 245 seconds)
17:01:04 Quit nick_p (Quit: Leaving)
17:02:03 Part Zagor
17:14:02 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:23:20 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:24:56 Join XavierGr [0] (~xavier@rockbox/staff/XavierGr)
17:29:02 Join kadoban_ [0] (
17:35:55 Join TheLemonMan [0] (
17:41:24 Quit enthdegree (Ping timeout: 248 seconds)
17:46:01 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
17:52:12 Quit remlap (Ping timeout: 245 seconds)
17:53:10 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
17:56:01 Quit mortalis (Quit: KVIrc 4.1.1 Equilibrium
18:11:14 Part LinusN
18:19:04 Quit bluebrother (Disconnected by services)
18:19:11 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
18:21:13 Quit fs-bluebot (Ping timeout: 246 seconds)
18:21:36 Quit T44 (Ping timeout: 276 seconds)
18:22:20 Join fs-bluebot [0] (
18:24:55 Join remlap [0] (
18:29:11 Join enth [0] (~BitchX@unaffiliated/enthdegree)
18:41:36***Saving seen data "./dancer.seen"
18:46:33 Quit tchan (Quit: WeeChat 0.3.7)
18:48:28 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:50:07 Quit enthdegree (Ping timeout: 245 seconds)
18:50:34 Quit enth (Ping timeout: 260 seconds)
18:53:30 Quit ender` (Quit: A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.-- Max Planck)
18:57:05 Join ender` [0] (
19:08:02 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:10:42 Quit benedikt93 (Quit: Bye ;))
19:11:53 Quit WalkGood (Quit: ♪ ♫ ♪ ♫ ♪ ♫ ♪)
19:38:07 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
19:53:37 Join lebellium [0] (
19:54:28 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
19:54:38 Join enth [0] (~BitchX@unaffiliated/enthdegree)
19:56:48*bertrik will have a closer look at the display init sequence for clip+ because of,33770.0.html
20:00:15 Quit XavierGr (Ping timeout: 244 seconds)
20:19:03 Join curtism [0] (
20:19:03 Quit curtism (Changing host)
20:19:03 Join curtism [0] (~curtis@unaffiliated/programble)
20:19:46 Join pamaury [0] (
20:19:46 Quit pamaury (Changing host)
20:19:46 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:22:39funmani'm looking to buy a nano2g in montreal (hand in hand, no post involved)
20:27:06gevaertsExcellent plan!
20:28:44 Join petur [0] (~petur@rockbox/developer/petur)
20:28:56 Join lovasoa [0] (~olojkine@
20:29:18kugelfunman: how many do you hide now?
20:36:40 Join Misanthropos [0] (
20:41:40***Saving seen data "./dancer.seen"
20:57:55 Quit pamaury (Ping timeout: 252 seconds)
21:02:18CIA-44Commit bb6ac6f in rockbox by 03Bertrik Sikken: sansa clip+: properly reset display using the reset pin
21:05:30CIA-44bb6ac6f build result: All green
21:07:41gevaertsbertrik: 3.11 material?
21:08:08bertrikpossibly, if it helps the person in the forum thread
21:11:17bertrikit's probably a good thing to go in anyway
21:12:55diginetfor someone who has reasonable experience programming, and a good understanding of hardware, how realistic would it be to try and port Rockbox to a new CPU architecture?
21:13:10diginet(specifically, the OpenRISC core from opencores)
21:14:05bertriksounds entirely possible
21:14:52bertrikit's 32-bit, right? how much RAM do systems with an OpenRISC core have typically?
21:15:03 Quit lovasoa (Remote host closed the connection)
21:15:17 Join pamaury [0] (
21:15:17 Quit pamaury (Changing host)
21:15:17 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
21:15:22diginetyeah OR is 32-bit, I'm not sure, but since it's a softcore as much as one would require I suppose
21:15:33diginetthe dev board comes with 64-mb IIRC
21:16:21n1sdiginet: is it supported by a free compiler, preferrably gcc?
21:16:24bertrikI think it's possible, but it probably would take quite a bit of time
21:16:46bertrikthe amount of time also depends a lot on the peripherals for display, storage, USB, etc.
21:16:48diginetit has definitely has gcc, fairly recent as well, I believ
21:16:51n1sif you have gcc support and enough ram i think a port should be feasible
21:17:25diginetthe peripherals for it are well documented, it uses an open standard bus called Wishbone
21:17:33n1sdiginet: so it's a fork of gcc, not supported by mainline?
21:17:47diginetI can't remember, let me look that up
21:18:32diginetI think it is a fork, it says 4.5.1 on their wiki
21:20:03n1sthat should be fine i think. the toolchain building script would need some tweaks to deal with it but that's hardly a blocker
21:20:24diginetdoe rockbox use uLibc? (sorry if that's a stupid question). I ask because OpenRISC also has that
21:20:31bertrikdiginet, do you have some more info about the board you're thinking about?
21:20:59n1sno, we have our own libc stuff, with bits from different places taken in as needed
21:21:04diginetah okay
21:21:20n1show fast is the chip?
21:21:39diginetwell, the OpenRISC is an FPGA core, so I'll probably be using some sort of dev board for an FPGA, rather than build my own.
21:21:45n1sand does it have fast memory, data cache, fast multiplier?
21:21:58diginetn1s, I believe it can go up to 250mhz on some of the newer FPGAs
21:22:49diginetthere's the wikipedia page, according to that it'ss upposed to be comparable to ARM10
21:23:07n1sso it's just as a hobby thing for yourself?
21:23:13diginetwell, sort of
21:24:00bertrikthe 250 MHz mentioned on the or1200 wiki page is for a hard core design, not a soft core I think
21:24:24 Quit y4n (Quit: Assumption is the mother of all fuckups)
21:24:30diginetI've had this goal of building a DAP suited to my specifications, since none I could find are. Sort of in the same vein as that Lyre project that stalled out. I went with the OR12k because it's open source
21:24:37diginetbertrik, oh, you might be right
21:25:10diginetwell, I think it should be able to get up to 100mhz at least. I've seen more complicated cores running that fast (i.e. the OpenSPARC)
21:26:02n1sdiginet: it would probably be way easier to use an arm based devboard
21:26:22n1sas rockbox already runs on arm
21:26:55diginettrue. I might just trying messing around with OpenRISC (I've been meaning to get an FPGA board regardless) and if it turns out to be too complicated, I'll just go with ARM
21:29:13diginetanyhow thank you guys for being such a help :)
21:32:08n1sthe wikipedia page says it has a harward arch, just how harward is it?
21:32:36diginetI think it's modifed harvard
21:32:58 Join wodz [0] (
21:33:03n1sso it should be able to load codecs from storage?
21:33:32wodzwith softcore you may try leon it performed better then openrisc when I was looking
21:33:53diginetwodz, oh? I didn't know that, I had forgotten about Leon
21:34:19diginetjust out of curiosity, why were you looking/comparing the two? (i.e. for what application)
21:35:05wodzdiginet: I read some phD thesis about performance of 32bit softcores
21:35:57diginetah okay
21:36:21wodzport for leon would be quite cool, this will leave only PPC port to have all major ones :-)
21:36:33diginetwell, SPARC is probably much better supported by gcc and what have you than OpenRISC
21:37:50wodzmay I ask for review of g#190 ?
21:37:51fs-bluebot3Gerrit review #190 at,190 : Move load_firmware() to separate file to share it between bootloaders and rolo(). by Marcin Bukat (changes/90/190/1)
21:39:42diginetyeah, the LEON core is much better, it's like half the size of OpenRISC
21:41:34diginetand it can run up to 150mhz according to the developer site
21:42:22n1sif it has a decent multiplication and memory performance that should be plenty
21:42:40diginetit's basically a SPARC8
21:42:46diginetand it does have an FPU
21:42:58diginetas well as some basic DSP
21:43:40n1srockbox doesn't use floating point (except in dome plugins for non time critical stuff)
21:44:13n1sdsp stuff for fixed point can come in handy though
21:44:35n1sespecially if the compiler can make good use of it
21:44:42 Quit pamaury (Ping timeout: 245 seconds)
21:45:53diginetoh right, duh. . .I knew that (from reading the Golden Quotes page, the one about WMA :P)
21:47:56n1swe do now have a fixed point wma decoder though
21:48:15diginetI love that quote though
21:48:24diginetdoes Rockbox have DTS support by any chance?
21:58:38 Quit wodz (Quit: Leaving)
22:00:45 Quit Thra11 (Ping timeout: 244 seconds)
22:00:53prof_wolfffhi, i want to upload 3 patches for review, patches 2 and 3 depends on patch 1, the patches are locally committed (3 commits), how can i send them to gerrit as three different patches/reviews?
22:01:15 Join bitcraft_ [0] (
22:01:48 Quit TheLemonMan (Quit: WeeChat 0.3.7)
22:02:07n1sdiginet: pretty sure we don't
22:02:55bertrikprof_wolfff, I thought stuff like that worked basically automatically
22:03:41bertrikdo we still have the git sand box?
22:04:23 Join bitcraf__ [0] (~bitcraft@
22:04:26prof_wolfffso "git push origin HEAD:refs/for/master" will upload all 3 patches?
22:04:42 Quit bitcraft (Ping timeout: 245 seconds)
22:04:54KiwiCamgevaerts: and [Saint]: I have managed to build the Linux Sim. I used these instructions in combination with this (The UiSim instructions are a little lacking. The first link got around that. I also was missing part of the SDL - libsdl1.2-dev which I duly installed.
22:06:43bertrikprof_wolfff, yes I think so
22:07:41 Quit bitcraft_ (Ping timeout: 252 seconds)
22:07:42prof_wolfffand then if i need to modify only one patch and upload a new version, can it be done using git −−amend?
22:07:49 Join lebellium_ [0] (
22:08:43 Quit lebellium (Ping timeout: 246 seconds)
22:08:45 Nick lebellium_ is now known as lebellium (
22:11:39bertrikprof_wolfff, I'm no git expert, but I think that won't work
22:12:15 Join Thra11 [0] (
22:12:16KiwiCamSo now I can try to get that %and function working. Question: I gather that in order to make a .zip file for my Clip, I update Git with "git pull −−rebase" (??) and then run the ../tools/configure again for the Clip+, choose N for Normal build, instead of S for Sim, then "make" and "make zip". Is that correct?
22:12:32bertrikif you changed a patch, you should commit and push it again. Gerrit recognises it by the Change-Id.
22:13:18bertrikKiwiCam, yes
22:14:02bertrikdepending on the number of cores in your system, you can add a -jX argument to make to speed up the build, where X is the number of cores
22:15:12bertrikKiwiCam, have you made a local change?
22:15:44KiwiCambertrik: Excellent. I'll make some notes so that I don't forget what I've done. Those instructions of MarcinBukat's were gold to me. I have four 4 cores on this machine.
22:16:05KiwiCambertrik: Local change?
22:16:38prof_wolfffbertrik: suppose i must do the changes on my local commit for that patch and then upload to Gerrit, will try to test in on the sand box, thanks!
22:17:39bertrikKiwiCam, did you modify the source code, I mean
22:18:33KiwiCamNo. I was just using Rasher's Sim under WINE. I never knew how to build the Linux Sim until now.
22:19:31KiwiCamFor the cores, I gather I use "make -j4" and "make -j4 zip". Correct?
22:20:27KiwiCamAt this stage, I really have no intention of changing the source code.
22:20:32bertrik-j4 is only really useful for the plain "make", not for "make zip"
22:20:50KiwiCamVery good.
22:20:58bertrikand using -jX is not necessary at al,, just a bit faster
22:21:23KiwiCamHandy to remember though. Thanks for the tip.
22:21:54bertrikoh, I usually use separate directories for each type of build, so for example I have a "build-clipzip" and a "build-clipzip-sim" directory
22:22:14kugelprof_wolfff: yes, use amend
22:22:34bertrikand use ccache
22:25:51KiwiCamI'd prefer to do that. So instead of "mkdir build-dir cd build-dir ../tools/configure" I do "mkdir build-clipplus cd build-clipplus ../tools/configure" & "mkdir build-clipplussim cd build-clipplussim ../tools/configure" or similar. Correct?
22:26:56prof_wolfffkugel: how can i use amend to modify an specific local commit (not the last commit)?
22:27:38kugelprof_wolfff: rebase −−interactive
22:30:04prof_wolfffkugel: thanks, will look at the −−interactive option
22:31:27bertrikchanging the middle patch in a set of three could be tricky, I don't know an easy way to do that
22:32:17bertrikcommit it on top of the others, then reinsert it as a fixup to the middle commit in interactive mode?
22:34:06kugeljust start the interactive rebase, pick a to amend for the specific commit. then do your changes and commit −−amend. finally finish the rebase with rebase −−continue
22:40:53 Quit petur (Quit: here today, gone tomorrow)
22:41:44***Saving seen data "./dancer.seen"
22:43:11lebelliumdear kugel, sorry for bothering you with that but when do you have time to review the FM patch? :)
22:44:12 Quit bitcraf__ (Remote host closed the connection)
22:44:44kugellebellium: I answered on FS
22:44:52kugelwaiting for lorenzo
22:45:26lebelliumOk did not see
22:45:29lebelliumI tell him now
22:47:42 Join bitcraft [0] (~bitcraft@
22:49:21 Join lorenzo92 [0] (~chatzilla@
22:51:43lorenzo92kugel: hey! well as I told you, i'm busy these days, so well, I tried some days ago to commit the patch to git but unfortunately I had troubles with certificate
22:51:53lorenzo92I should retry the whole process
22:52:07kugelthere is no certificate involved :)
22:52:20lorenzo92uhm...well no, sorry, the personal key!
22:52:34lorenzo92or how's called...
22:52:53kugelyou need to add your ssh public key to your gerrit profile
22:53:23lorenzo92yes, i did this, so probably I'd better create a new key. maybe tomorrow I will have some spare time to work on that
22:54:05lorenzo92anyway, going back to the module, it should be possible to compile with our toolchain
22:54:11lorenzo92what do you think?
22:54:27kugelI didn't manage to compile it yet
22:54:38kugelhow do you do it?
22:55:50lorenzo92I put it simply in the samsung's kernel, then I compiled it as a module (via menu-config). But wait, if you can, take a look at the driver's package (where afe.h, r0gpio.h headers are)
22:56:38lorenzo92this the makefile used by samusng
22:57:01kugeladding a module to the kernel is not that simple :)
22:57:47lorenzo92well, I simply followed the other modules, changing it's name :D
22:59:25 Join Keripo [0] (
22:59:40lorenzo92kugel: yeah searching in google for "compile kernel module via makefile" gives some interesting results, possibly
23:03:02 Part TomColler
23:06:36kugellorenzo92: couldn't be bothered with that, sorry
23:06:42 Quit Keripo (Quit: Leaving.)
23:06:56kugelI hoped you can give some recipe, or even better code/patch to compile :)
23:07:23lorenzo92kugel: okay, I will try to do the best :)
23:07:48lorenzo92first of all trying to build it without kernel source
23:08:29kugelit should work with only kernel headers (make headers_install) but that was troublesome
23:19:11 Quit n1s (Read error: Connection reset by peer)
23:22:47 Quit lorenzo92 (Remote host closed the connection)
23:31:26 Join dhrasmus [0] (
23:33:15 Join remlap1 [0] (
23:33:46 Quit remlap1 (Client Quit)
23:34:27 Join jhMikeS [0] (
23:34:27 Quit jhMikeS (Changing host)
23:34:27 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
23:35:20 Quit remlap (Ping timeout: 252 seconds)
23:52:56 Quit lebellium (Quit: ChatZilla [Firefox 12.0/20120314195616])

Previous day | Next day