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

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

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

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

#rockbox log for 2014-07-22

00:04:47 Join RiD [0] (
00:06:37 Quit AKBlindy (Ping timeout: 240 seconds)
00:09:51 Quit the-kyle (Ping timeout: 240 seconds)
00:11:35 Quit TheSeven (Ping timeout: 260 seconds)
00:12:11 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
00:13:16 Join AKBlindy [0] (
00:21:22 Join the-kyle [0] (
00:28:56 Join RiDD [0] (
00:30:42 Quit RiD (Ping timeout: 240 seconds)
00:44:21 Quit the-kyle (Ping timeout: 264 seconds)
00:52:27***Saving seen data "./dancer.seen"
00:54:42 Quit ZincAlloy (Quit: Leaving.)
01:06:35 Quit ender` (Quit: Cigarettes are like squirrels. They're perfectly harmless until you put one in your mouth and light it on fire.)
01:10:48 Join the-kyle [0] (
01:10:59 Quit Strife89 (Ping timeout: 256 seconds)
01:19:34 Quit Naked (Remote host closed the connection)
01:38:10 Quit GodEater (Ping timeout: 250 seconds)
01:38:22 Join GodEater [0] (
01:38:22 Quit GodEater (Changing host)
01:38:22 Join GodEater [0] (~whoknows@rockbox/staff/GodEater)
01:39:33 Join sakax [0] (~sakax@unaffiliated/sakax)
01:46:10 Quit bertrik (Remote host closed the connection)
02:13:37 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
02:17:27 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
02:23:47 Join skx_netb [0] (~sakax@unaffiliated/sakax)
02:30:44 Quit sakax (Quit: Leaving)
02:34:09 Quit cmhobbs (Ping timeout: 264 seconds)
02:40:15[Saint]TheDarkPirate: As bertrik alluded to earlier, ratings in Rockbox are entirely non-obvious. For some reason that I am completely unaware of, we use a proprietary rating system instead of using standardized embedded metadata.
02:40:48[Saint]So the first step in sorting by ratings in Rockbox, is to painstakingly rate each file by hand from the wps context menu.
02:41:48 Quit [Saint] (Quit: going down for a system upgrade)
02:42:12 Join [Saint] [0] (~saint@rockbox/staff/saint)
02:42:51 Part [Saint]
02:43:46 Join [Saint] [0] (~saint@rockbox/staff/saint)
02:44:37 Part [Saint]
02:46:23 Join [Saint] [0] (~saint@rockbox/staff/saint)
02:47:33 Quit [Saint] (Quit: Quit.)
02:50:22 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
02:52:30***Saving seen data "./dancer.seen"
02:59:11 Join [Saint] [0] (~saint@rockbox/staff/saint)
03:00:00 Quit AlexP (Remote host closed the connection)
03:36:33 Join ikeboy [0] (
03:40:41 Quit ikeboy (Ping timeout: 245 seconds)
03:43:48 Quit RiDD (Quit: A good plan today is better than a perfect plan tomorrow.)
03:53:05 Quit skx_netb (Remote host closed the connection)
04:10:54 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:10:54 Quit amiconn (Disconnected by services)
04:10:54 Quit pixelma (Disconnected by services)
04:10:55 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:10:57 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:10:57 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:19:19 Join ygrek [0] (~user@
04:23:26 Join ikeboy [0] (
04:26:32 Quit ikeboy (Client Quit)
04:48:19 Quit Cinos (Ping timeout: 240 seconds)
04:52:31***Saving seen data "./dancer.seen"
04:53:01 Join Cinos [0] (
05:36:27 Join steffengy [0] (
05:39:26 Quit steffengy1 (Ping timeout: 245 seconds)
05:41:40 Join kugel [0] (~kugel@rockbox/developer/kugel)
05:53:24 Quit TheSeven (Ping timeout: 250 seconds)
05:55:18 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:08:48 Join HeadWhacker [0] (7738794d@gateway/web/freenode/ip.
06:10:27HeadWhackerhi, is it possible to capture a keycode when pressing a button from a rockbox log?
06:11:18JdGordonwhat are you trying to do?
06:14:15HeadWhackerI was able to make progress with Rockbox on DX90. All physical buttons are working except the prev (<<) button. If I can capture the keycode integer while pressing the button then I can fix it.
06:18:24JdGordonhave you got any logging options on the device?
06:18:44JdGordonmaybe fix the debug gpio's screen for it and press buttons and see what values change
06:26:57 Quit HeadWhacker (Ping timeout: 246 seconds)
06:31:43 Quit ygrek (Ping timeout: 240 seconds)
06:36:34 Join telliott [0] (
06:52:34***Saving seen data "./dancer.seen"
07:30:14 Join mortalis [0] (~kvirc@
08:14:51 Quit GodEater (Ping timeout: 240 seconds)
08:16:31 Join GodEater [0] (
08:16:31 Quit GodEater (Changing host)
08:16:31 Join GodEater [0] (~whoknows@rockbox/staff/GodEater)
08:21:35 Join ender` [0] (
08:24:24 Quit telliott (Read error: Connection reset by peer)
08:25:20 Join ygrek [0] (~user@
08:31:45 Quit mortalis (Ping timeout: 264 seconds)
08:31:51 Nick DormantBrain is now known as SuperBrainAK (~andy@
08:52:35***Saving seen data "./dancer.seen"
08:54:43 Quit GodEater (Ping timeout: 240 seconds)
09:00:48 Join einhirn [0] (
09:08:31 Quit ygrek (Remote host closed the connection)
09:08:47 Join ygrek [0] (~user@
09:15:08 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:17:07 Join mortalis [0] (~kvirc@
09:22:19pamauryTheSeven: I saw your comment about the SCSI code, indeed it's not very robust
09:28:03 Nick SuperBrainAK is now known as DormantBrain (~andy@
09:47:35 Join LinusN [0] (
10:04:43 Join tfx [0] (c1afdffd@gateway/web/freenode/ip.
10:04:49tfxhi @ all
10:04:57tfxi want ot ask for support
10:05:03tfxand help
10:06:09tfxhow can i create the speech files within the rockbox utility: where can i get the flite.exe
10:06:12pamaurytfx: ask your question, if anyone can help he/she will answer
10:06:50tfxor should i use another speech engine to create voice files?
10:06:52pamaurythe installer support various speech encoder iirc, you can setup the one you prefer in the preferences, of course if you want to use flite you have to install it before
10:08:30tfxthe problem is - i don't know where i can get the flite files. want to use sapi5 before, but it hangs every time during the creation
10:09:45tfxof the speech files
10:10:14tfxcould some kindly share some download links for the binary files for windows
10:11:14pamauryI don't know, I guess you have to install it, I don't run windows
10:13:03tfxok then i'll try to compile this
10:13:43pamauryyeah looking on google, there don't seem to be an easy to find prebuilt version
10:14:52tfxyes indeed, i've searched all over before
10:15:11tfxfound an prebiult flite version for italian language only
10:16:34tfxbut i'll try to compile the files - so i don't need to search over and over
10:17:28tfxif you wish i can afterwards share the prebuilt file for other users
10:17:53tfxthe next problem occurs while playing some vgm files: the sound is shuttering on my iriver h10 5gb ums device
10:18:11tfxis this 'bug' already known
10:20:06pamauryI don't know, it is possible that the iriver is not powerful enought to decode it ini realtime
10:20:26tfxis a patch or another solution possible?
10:20:26pamaurybut I don't own the device so I'm not too aware of the known bugs for it
10:21:38pamauryif you are not afraid of compiling you could try to compile the sim and play it, see if the stutter is related to the codec or the device
10:22:00pamauryor do you have another device to test on ?
10:22:11tfxsorry no
10:22:36pamauryok, can you upload it ? I can try on my device
10:23:03tfxcould i use the prebuild sim from too?
10:24:46tfxthanks - so i'll try it
10:30:02 Quit Rower (Ping timeout: 240 seconds)
10:34:02tfxthe vgmz playback with the sim from is accurate, no shuttering occured
10:35:52tfx@ pamaury i used some vgmz files from for thsi test i've used the dune - battle for arrakis music
10:36:34 Join shdwprince [0] (~sp@
10:38:45pamaurytfx: so maybe it's because your device is not fast enough
10:39:37tfxcan i adjust some settings or everything else to try
10:40:39tfxi remember with an *old* rockbox version these files play very accurate and without shuttering
10:41:27tfxcurerntly the shuttering occures ever three seconds
10:41:39pamauryhum, so that's a regression indeed
10:42:03tfxit's like someone pressed pause every three seconds and then resume
10:42:04pamaurylet me try on my fuze+
10:42:12pamauryany of these file will stutter on yours ?
10:42:38tfxi tried to unpack the vgz files already but without luck
10:42:52tfxthe vgm files shutters too
10:44:33[Saint]I just got some really fucking clever variable state checking implemented...aaaaaaaaaaaaaaaaaannnnnnnd, it draws a fat black line over my album art.
10:45:34pamaurytfx: it works smoothly on my fuze+, so apparently that's a iriver regression
10:45:39[Saint]And, I have absolutely *no* idea what line(s) {is|are} causing it.
10:45:43pamaurywe need to find a dev who has this device
10:45:56[Saint]"back in the day" I used to be able to push this effect offscreen with newlines.
10:46:08[Saint]Which was an ugly hack/workwaround, but, it worked.
10:46:28[Saint]Every corner I turn with this bloody theme I hit a wall.
10:46:59[Saint]10 implement_something
10:47:06[Saint]20 break_something
10:47:10[Saint]30 goto 10
10:47:30pixelmapamaury: the Iriver H10 is more like an old Ipod or Sansa c200/e200 - portalplayer
10:47:49 Quit shdwprince (Ping timeout: 255 seconds)
10:49:57*[Saint] wonders if he can cheat and use this conditional state checking to re-draw the album art
10:50:01tfxok currently i tried the playback with the latest dev release, but the same shuttering here
10:50:09[Saint], that's an ugly hack.
10:50:11pamaurytfx: you could try to bisect
10:50:23[Saint]This was supposed to be _simple_. :-/
10:51:14pamaurytfx: you take a known working revision and using a tool called git bisect, you build and try ~15 builds to determine which commit broke playback
10:51:54pamaurythat's a lengthy process but that's pretty much the only option
10:52:37***Saving seen data "./dancer.seen"
10:54:42tfxphew - compiling, git, check errors ... all for rockbox
10:56:48tfxi'll try my best
10:56:54pamaurytfx: thanks a lot
10:56:57tfxbut please give me some time
10:57:11tfxi need to determine some free time for this learning curve first
10:58:58TheSevenpamaury: what's your opinion on adding ATA passthrough / SMART to rockbox?
10:59:26TheSevenit would need some additional interfacing between the SCSI command interpreter and ATA driver
10:59:42pamauryTheSeven: I think that would be cool, to be honest I don't remember much of that part of the specification though
11:00:20tfxbtw there are no icons set for - *.cells *.z5 could someone please fix this for the next releases
11:00:49TheSevenpamaury: we basically need thos:[]=/@965&compare[]=/@967
11:01:53TheSeventhat patch should also fix most of the cases where the host expects to transfer more data than we do, which is typical for nonimplemented commands
11:02:29TheSevenI didn't take care of the other cases (host expects to transfer less than we do) yet, those will still lock up the driver
11:02:56TheSevenor wait, actually I catch those and crash the driver to pinpoint that the cause
11:06:39TheSevennot sure how much emcore's disk mode code and rockbox's scsi code differ these days though
11:06:52TheSevenso porting it might be a bit more complicated than it seems
11:16:39[Saint](still lots of dummy assets)
11:16:56[Saint](...and lots of unimplemented assets)
11:17:31 Quit einhirn (Ping timeout: 260 seconds)
11:17:32[Saint]Just got to finish working on the overflow menu.
11:17:53[Saint](that's where everything you can't see in the wps is hidinh)
11:18:20[Saint]I wanted to keep it simple, less visually distracting.
11:18:37[Saint]Only give the user the bare minimum of what they need to see at any one point in time.
11:19:26[Saint]I also want to keep a "rockboxesque" feel to it.
11:19:39[Saint]Hence the yellow and light grey smattered around the place.
11:19:56tfxwow look sgoot - please port to irevier h10 5gb
11:20:28tfxwhen finished
11:21:51[Saint]That's the intention. I want to get it done for all the targets where it is feasible to do so, but my main focus at the moment is touchscreen.
11:23:00[Saint]I'm using various techniques that _should_ allow for this theme to be trivially ported to arbitrary resolutions and be largely resolution-agnostic.
11:24:32JdGordon[Saint]: can you use the % options that ploco just added?
11:24:49[Saint]In some places, yes, indeed I can.
11:25:05[Saint]% combined with negative offset placement
11:25:25[Saint](and, the key top it all, no backdrop)
11:26:01[Saint]Once you add a backdrop you imply a very specific resolution (which I don't have to tell you), so obviously I'm avoiding that.
11:28:06[Saint]But...anyhoo, that's what she's lookin' like so far. I want to keep the overflow menu hidden as a bit of a surprise, because I think people are going to be blown away by it.
11:28:33[Saint](well, perhaps "blown away" is a bit far...but, I happen to think its pretty bloody clever :))
11:30:17tfxplease share your code (when it's done)
11:31:15[Saint]Oh, I will. Don't worry about that. :)
11:31:25tfxi still *need* to update my old themes with fms and multi-lingual support and what i saw from your screenshat was very detailed commented
11:31:37[Saint]Once I *finally* finish this thing, I'll be making a big fuss about it.
11:31:47tfxand yes i'll include credits
11:32:13tfxwill this info @ too?
11:32:25*[Saint] nods
11:33:25tfxand will this the new *default* rockbox theme?
11:33:47tfxjust like cabbie?
11:34:12tfxis this still the default one? not sure?
11:35:06[Saint]I'm hoping to replace the default theme for the hosted application targets (SDL, Android, Maemo, etc.). Not sure about the native (DAP firmware replacement) targets as yet.
11:35:33[Saint]That would be great, but its a big job.
11:37:16[Saint]There's about 40 different variants of cabbie presently, but hopefully this work enables us to reduce that number drastically.
11:46:03tfxyes this would be great. i remember for the creation of my *modded* themes i downloaded more than then other themes only to inspect the code and found more than one solution possible to eg view alternate titles or playpos bar but there was no *default* solution. currently i want to add the viewpoint-divs but i don't know if this is the *default* way of rockbox
11:50:27 Join AlexP [0] (~alex@rockbox/staff/AlexP)
11:53:19[Saint]Yes. One of the purposes of this theme is to (hopefully) serve as a canonical example of syntax usage
11:53:49[Saint]That's one of the reasons why it is divided into separate elements and heavily commented.
11:54:46 Quit copper (*.net *.split)
11:54:46 Quit charlie (*.net *.split)
11:54:46 Quit olspookishmagus (*.net *.split)
11:54:47 Quit Kohlrabi (*.net *.split)
11:54:48 Quit GeekShadow (*.net *.split)
11:54:48 Quit Bluefoxicy (*.net *.split)
11:54:48 Quit n17ikh (*.net *.split)
11:54:48 Quit Makinit (*.net *.split)
11:54:51[Saint]But the primary reason for that, honestly, is me being particularly anal about syntax and coments. ;)
11:54:54 Join Makinit [0] (
11:54:55 Join olspookishmagus [0] (
11:54:55 Join Kohlrabi [0] (
11:55:01 Join Bluefoxicy [0] (
11:55:03 Join n17ikh [0] (~n17ikh@
11:55:03 Quit n17ikh (Changing host)
11:55:03 Join n17ikh [0] (~n17ikh@unaffiliated/n17ikh)
11:55:19 Nick olspookishmagus is now known as Guest1082 (
11:55:23[Saint]My themes have always been generously sprinkled with comments and nicely formatted stylistically.
11:55:30[Saint](well, IMO at least)
11:56:06 Join GeekShadow [0] (
11:56:06 Quit GeekShadow (Changing host)
11:56:06 Join GeekShadow [0] (~antoine@reactos/tester/GeekShadow)
11:57:01 Join copper [0] (~copper@unaffiliated/copper)
11:57:18 Join shdwprince [0] (~sp@
11:57:32 Join charlie [0] (
12:14:18 Join RiD [0] (~RiD@
12:23:18 Join ulmutul [0] (
12:23:45ulmutultfx: still there?
12:25:11ulmutulI just read the logs. I remember I had stuttering playback with the peakmeter enabled.
12:26:09ulmutulThe cabbie theme wasn't released yet for the target, and the failsafe's peakmeter made the playback stutter.
12:26:31tfxso i need to diasble the peakmeter?
12:26:36ulmutulSo you may check your theme before you begin to bisect.
12:26:59ulmutulTry a theme (e.g. cabbie) without peakmeter.
12:27:57ulmutulOr just wait a few seconds. Because on my target the stuttering disappeared when the lcd switched to sleep mode :)
12:28:10tfxcurrently im using this theme
12:28:59tfxno peakmeter there but i'll try the sleep mode
12:29:03tfxwhile playing
12:30:30[Saint]ulmutul: thank you for reminding me
12:30:44[Saint]I need to remove the peak meter from the fallback theme
12:30:55[Saint]It makes absolutely zero sense being there
12:32:10[Saint](for the reason you stated)
12:32:24ulmutulEspecially because it's named "failsafe" :)
12:33:04tfxbetter than a blank screen or anything else
12:33:37[Saint]The peak meters are *very* computationally intensive.
12:33:55[Saint]It makes zero sense for them to be there on a fallback theme.
12:34:55[Saint]Battery life improved around 45 minutes when I benched fallback without peak meters vs. fallback with peak meters.
12:35:02[Saint]That's quite a lot.
12:41:48shdwprinceI wonder, how to debug-print in simulator build? printf don't work, logf too.
12:43:04shdwprinceI'm shure that code executes, but neither printf or logf dont pring anything.
12:44:22shdwprinceOh, thanks, works well
12:52:38***Saving seen data "./dancer.seen"
12:54:18 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
12:59:38[Saint]Man I love high resolution targets.
12:59:54[Saint]I have enough room for 15 lines of 40px metadata
13:00:06[Saint]That's a whole lotta' metadata, yo.
13:18:53 Join ZincAlloy [0] (
13:26:51 Quit ygrek (Ping timeout: 240 seconds)
13:30:39 Quit tfx (Quit: Page closed)
14:29:32 Quit shdwprince (Quit: WeeChat 0.4.3)
14:36:11 Join einhirn [0] (
14:43:24 Join telliott [0] (
14:47:56 Quit telliott (Client Quit)
14:52:42***Saving seen data "./dancer.seen"
14:53:52 Quit cmhobbs (Ping timeout: 250 seconds)
14:58:11 Join telliott [0] (
14:58:12 Join amayer [0] (
15:01:29 Quit telliott (Client Quit)
15:01:51 Join telliott [0] (
15:04:42 Quit telliott (Client Quit)
15:07:03 Join telliott [0] (
15:10:32 Join ygrek [0] (~user@
15:15:52 Part TheDarkPirate
15:23:09 Quit telliott (Read error: Connection timed out)
15:58:38 Part LinusN
15:59:40 Quit ranmachan (Ping timeout: 255 seconds)
16:03:11 Join ikeboy [0] (
16:07:27 Quit pamaury (Ping timeout: 256 seconds)
16:10:06 Join sakax [0] (~sakax@unaffiliated/sakax)
16:10:46copperI've confirmed with someone else that the recent Opus speed improvements produce very decent battery life on a Clip+
16:11:55 Quit sakax (Client Quit)
16:12:19 Join sakax [0] (~sakax@unaffiliated/sakax)
16:14:00 Quit jhMikeS (Quit: Confucius say: The short dandelion survive the lawnmower)
16:14:33 Quit Jinx (Quit: reboot)
16:21:39 Join pretty_function [0] (~sigBART@
16:29:23 Join jhMikeS [0] (~jethead71@
16:29:23 Quit jhMikeS (Changing host)
16:29:23 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
16:30:19 Quit kugel (Quit: Lost terminal)
16:38:46 Quit mortalis (Quit: KVIrc 4.3.1 Aria
16:52:46***Saving seen data "./dancer.seen"
17:10:43 Quit sakax (Remote host closed the connection)
17:10:54 Quit bertrik (Ping timeout: 272 seconds)
17:34:58 Join SolarAquarion [0] (~SolarAqua@unaffiliated/solaraquarion)
17:35:10SolarAquariongevaerts, hi
17:35:27SolarAquarioni installed rockbox on my ipod so how do i add music and what not to it
17:35:40SolarAquariondo i use any music thing?
17:36:31SolarAquarioni need some support moving forward if anyone can help
17:38:45copperSolarAquarion: you connect your iPod to your computer with the USB cable, and copy files onto it like you would with a regular USB drive
17:39:07SolarAquarioncopper, thanks and rockbox categorizes them?
17:39:10SolarAquarionor whatever
17:39:16copperthat would be the databse
17:39:33coppergo in the database settings and select "Initialize" the first time
17:39:53copperthen you can manually update the database or you can set it to update itself automatically
17:40:00SolarAquarionwhat kind of videos does rockbox support
17:40:14copperdunno, check the manual or the wiki
17:42:01SolarAquarionjust regular mpeg files dOK]
17:44:31SolarAquarionrockbox is awesome
17:50:15copperIt's not awful.
17:51:25 Join syscrash [0] (
17:51:25 Quit syscrash (Changing host)
17:51:25 Join syscrash [0] (~syscrash@poipu/developer/syscrash)
17:53:30 Nick syscrash is now known as jvd (~syscrash@poipu/developer/syscrash)
17:53:34SolarAquarioncopper, it's better for linux stuff and that you don't need to fuck with database stuff on linux
17:56:49 Part jvd
17:59:42 Join Rower [0] (
18:01:35 Quit ikeboy (Quit: Leaving)
18:13:37 Quit pretty_function (Quit: Leaving...)
18:16:13 Quit jhMikeS (Ping timeout: 240 seconds)
18:24:10 Join ikeboy [0] (
18:41:47 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:42:30 Join GodEater [0] (
18:42:30 Quit GodEater (Changing host)
18:42:30 Join GodEater [0] (~whoknows@rockbox/staff/GodEater)
18:45:25 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
18:52:49***Saving seen data "./dancer.seen"
18:54:43 Quit ygrek (Ping timeout: 255 seconds)
18:59:15 Quit ikeboy (Quit: Leaving)
19:11:03 Join ygrek [0] (~user@
19:30:24 Join lebellium [0] (
19:35:25 Quit fs-bluebot (Ping timeout: 250 seconds)
19:38:43 Join fs-bluebot [0] (
19:46:15 Quit rudi_s (Ping timeout: 250 seconds)
19:49:29 Join rudi_s [0] (
19:54:11 Join pretty_function [0] (~sigBART@
20:11:26 Quit pamaury (Read error: Connection reset by peer)
20:12:14 Quit Rower (Ping timeout: 255 seconds)
20:12:33 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:28:56 Quit Xerion (Quit: )
20:35:16 Join wodz [0] (
20:35:24wodzpamaury: ping
20:35:31pamaurywodz: pong
20:36:17pamauryyou want the YP-CP3 photos ? Let me check if I still have them...
20:36:31wodzpamaury: I can't remember how to patch mem location with hwpatcher, could you help? I need to patch jumptable
20:36:40lebelliumI'm interested too :P I never dared to disassemble mine
20:37:08 Nick Guest1082 is now known as olspookishmagus (
20:37:16pamaurywodz: let me have a look, I don't remember :p
20:37:54wodzpamaury: YP-CP3 firmware is strange. It doesn't work when loaded with rk27load. It does some strange things on init like setting SP to 0x60e00000 which is out of physical ram afaik
20:38:10pamauryhave a look at the others patches, assuming you have loaded the fw, you can write any location like this:
20:38:10pamauryhwp.write32(fw, addr, value)
20:38:23wodzah, goodie
20:38:24pamauryand the "addr" must the created with hwp.make_addr
20:38:31wodzyes I remember
20:40:25wodzpamaury: photos would be nice, yes
20:40:45pamaurywodz: unfortunately I cannot find them
20:40:50pamauryI must have trashed them by mistake
20:41:02wodzno problem
20:41:07pamauryusually I archive all the photos of my targets but for some reason I don't have them in the expected place
20:44:50wodzyeah, as expected SP=0x60e00000 doesn't work on rk27load'ed firmware. It crashes on jump to main(). Now why this is valid when loaded from bootloader?
20:47:08wodzpamaury: do you have any clue?
20:47:37pamaurywodz: doesn't the stub setup his own stack ?
20:48:42 Quit ygrek (Ping timeout: 260 seconds)
20:48:54wodzpamaury: yes, it does and that is working fine. What I am trying to do is to load OF with rk27load. This crashes just at jumping to main()
20:49:51wodzpamaury: If I patch main address in trampoline function to jump into hwstub directly it works, if I patch first instruction of main() to jump to hwstub it doesn't
20:50:22 Join v1 [0] (~v@
20:51:06v1hi everyone, does anyone knows how to start rockbox on arch linux? (it's already installed)
20:51:39wodzv1: what you mean by starting rockbox on arch linux? sim or what?
20:52:32v1yes startint rockbox, what do you mean by sim?
20:52:49pamaurywodz: does the OF self-relocate ?
20:52:50***Saving seen data "./dancer.seen"
20:52:59pamaurynot sure that's important though
20:53:30pamaurywodz: maybe there is a problem with the instruction patching, did you try to disassemble it to check it's correct
20:53:54the-kylev1: Are you trying to start the Rockbox Utility? What desktop environment are you using?
20:54:10wodzpamaury: 1) AFAIK OF doesn't self relocate 2) Yes I checked patched binary and looks ok
20:54:13v1I am on openbox
20:54:53the-kylev1: Try running RockboxUtility.
20:55:01pamaurywodz: I suggest you disassemble it by hand, I ran into an issue recently when the instruction *looked* correct but it was *unpredictable* on some ARM cores
20:55:09pamauryno disassembler was aware of it
20:55:24v1ok tab was evidently not working with rock nor rb, thanks a million
20:55:36the-kyleNo prob.
20:55:43v1that workedout perfectly!
20:56:07wodzpamaury: what you mean 'by hand'? Both objdump and ida agree about this jump
20:57:19pamauryI mean use the ARM manual to decode it yourself
20:57:24pamauryI know but it proves nothing
20:57:44pamauryespecially for blx instructions, there are corner cases
20:58:12wodzpamaury: this is straight unconditional b instruction
20:58:47pamauryin thumb or arm mode ?
20:59:53 Join ranmachan [0] (
21:00:54pamaurylooks pretty armless indeed (pun itended)
21:01:26wodzanyway why OF sets SP to invalid address ?
21:01:35pamauryindeed that's very confusing
21:01:47wodzTheSeven: there is no mmu on this core
21:01:59pamaurydid you check it's actually invalid ? isn't there some kind of mirroring so that it's finally decoded to RAM ?
21:02:32 Join ikeboy [0] (
21:03:38wodzpamaury: Yes it should be mirrored. I assumed it will end up at the beginning of ram after address wrap but it may be not the case actually
21:04:15TheSeventhis kind of thing is sometimes used to provide uncached aliases of memory regions on targets without a MMU
21:05:08wodzhmm, it is not unthinkable indeed
21:05:26wodzbut uncachable stack?
21:05:50TheSevenseems weird, yes
21:06:09TheSevenbut no idea what they're doing, so I'm just suggesting things ;)
21:06:42wodzno, unmap regions are setup such that this address is treated as cached
21:07:38 Quit v1 (Quit: WeeChat 0.4.3)
21:08:12 Join cmhobbs_ [0] (
21:17:02 Quit pretty_function (Remote host closed the connection)
21:35:07 Join Xerion [0] (
22:01:31 Quit dfkt (Remote host closed the connection)
22:05:41 Quit ikeboy (Quit: Leaving)
22:06:06 Join dfkt [0] (OxO29A@unaffiliated/dfkt)
22:31:39 Quit Provel (Ping timeout: 256 seconds)
22:32:07 Join Provel [0] (
22:36:18 Quit ZincAlloy (Quit: Leaving.)
22:44:45wodzHmm, rkw header refers to address higher then dram size I assumed. Maybe dram init routine is not valid for this setup?
22:52:54***Saving seen data "./dancer.seen"
23:18:24 Quit amayer (Quit: Leaving)
23:19:57 Join Strife89 [0] (
23:32:29 Quit pamaury (Ping timeout: 255 seconds)
23:37:02 Quit lebellium (Quit: ChatZilla [Firefox 31.0/20140716183446])
23:54:19 Quit TheSeven (Ping timeout: 272 seconds)
23:55:53 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
23:58:35 Quit GodEater (Ping timeout: 245 seconds)

Previous day | Next day