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 2012-09-18

00:04:42 Nick LittleCreature is now known as FearOfMusic (~fearofmus@unaffiliated/fearofmusic)
00:06:04pamaury[Saint]: ok, nice to know :)
00:12:30 Quit Totalled (Quit: PETTAN PETTAN, TSURUPETTAN!)
00:12:37 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
00:19:59 Join Totalled [0] (
00:22:08 Quit pamaury (Ping timeout: 252 seconds)
00:27:31 Quit [7] (Disconnected by services)
00:27:40 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
00:34:27 Quit kevku (Ping timeout: 260 seconds)
00:44:27 Quit factor (Read error: Connection reset by peer)
00:54:05 Quit bertrik (Remote host closed the connection)
00:57:16 Quit eckoit (Quit: eckoit)
00:58:42 Quit ack (Ping timeout: 244 seconds)
01:05:55 Quit mgottschlag (Ping timeout: 240 seconds)
01:08:41 Quit AlexP (Ping timeout: 272 seconds)
01:10:09 Quit Torne (Quit: Lost terminal)
01:10:44 Join Torne [0] (~torne@rockbox/developer/Torne)
01:12:17 Join eckoit [0] (~ryan@
01:17:49 Join factor [0] (
01:19:52 Join AlexP [0] (~alex@rockbox/staff/AlexP)
01:46:30***Saving seen data "./dancer.seen"
02:16:28 Join entombed [0] (3ffd71c2@gateway/web/freenode/ip.
02:17:49 Part entombed
02:50:15 Join davo [0] (
03:07:15 Quit Jack87 (Ping timeout: 240 seconds)
03:11:32 Join CIA-16 [0] (
03:14:03 Join Jack87 [0] (Jack87@nasadmin/admin/jack87)
03:16:03[Saint]what package does freetype-config come from?
03:19:19[Saint]aha, foundeded it.
03:22:13 Join amayer [0] (
03:46:34***Saving seen data "./dancer.seen"
04:13:07 Join perrikwp_ [0] (
04:13:38 Quit perrikwp (Ping timeout: 252 seconds)
04:30:33 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:30:33 Quit pixelma (Disconnected by services)
04:30:33 Quit amiconn (Disconnected by services)
04:30:34 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:30:36 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:30:36 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:36:20 Part amayer
04:38:02 Join zz_TheSphinX^ [0] (
04:40:53 Quit guymann (Changing host)
04:40:53 Join guymann [0] (~c@unaffiliated/guymann)
04:41:48 Quit TheSphin_ (Ping timeout: 245 seconds)
04:43:48 Join entombed [0] (d1b4f629@gateway/web/freenode/ip.
04:55:19 Part entombed
05:04:30 Nick FearOfMusic is now known as LittleCreature (~fearofmus@unaffiliated/fearofmusic)
05:08:23 Quit TheSeven (Disconnected by services)
05:08:32 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:15:33[Saint]hmmmm, building the chinese-traditional .voice file takes ~250% longer than english
05:15:55[Saint]hindi is also quite a slow one, it seems.
05:22:14 Join entombed [0] (d1b4f629@gateway/web/freenode/ip.
05:25:05 Join Rower85 [0] (
05:33:16 Quit ps-auxw (Ping timeout: 240 seconds)
05:33:25 Join kingjames [0] (43530b66@gateway/web/freenode/ip.
05:33:36 Part kingjames
05:35:35 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
05:46:36***Saving seen data "./dancer.seen"
05:47:55 Quit entombed (Ping timeout: 245 seconds)
05:52:39 Quit eckoit (Quit: eckoit)
05:57:21[Saint]the version that Rockbox displays during that hardcoded somewhere during compile time, or just pulled from rockbox-info.txt?
06:07:47Tornehardcoded at compile time
06:08:59[Saint]right, so, rockbox-info.txt (unsurprisingly) is just for Rockbox Info in the System settings?
06:09:21[Saint]I seem to recall there's a way to pass an arbitrary version styring at compile tiome, no?
06:09:35[Saint]dammit! *string
06:11:44 Join eckoit [0] (~ryan@
06:14:10[Saint]Torne: you don't happen to know that one, do you? I seem to recall it being possible.
06:14:20[Saint]Some compile time flag can do it, iirc.
06:16:02Tornemake VERSION=foo i think
06:16:19[Saint]thank you.
06:16:25[Saint]I'll give it a go.
06:19:12[Saint]Yay, that seems to have done the trick. Thank you Torne. I tried to look it up in the source but I got lost.
06:24:58 Quit Scromple (Ping timeout: 256 seconds)
06:25:01[Saint]hmmmm, maybe not.
06:35:50 Join Scromple [0] (~Simon@
07:14:41bluebrother[Saint]: see
07:15:04[Saint]bluebrother: yeah, thanks, I figured it out.
07:28:01 Quit davo (Quit: Lost terminal)
07:40:52 Quit bluebrother (Disconnected by services)
07:40:56 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
07:43:16 Quit fs-bluebot (Ping timeout: 272 seconds)
07:44:43 Join fs-bluebot [0] (
07:46:39***Saving seen data "./dancer.seen"
07:56:11 Join Keripo [0] (
08:01:25 Join XavierGr [0] (~xavier@rockbox/staff/XavierGr)
08:15:07 Quit Keripo (Read error: Connection reset by peer)
08:42:20 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
08:43:37 Nick [Saint] is now known as notevilnick (~saint@rockbox/user/saint)
08:43:42 Nick notevilnick is now known as [Saint] (~saint@rockbox/user/saint)
08:56:01 Join ender` [0] (
08:56:28 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:58:27 Quit mgottschlag (Remote host closed the connection)
09:00:11 Join kevku [0] (
09:05:00 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
09:05:33 Join LinusN [0] (
09:06:37 Join megal0maniac [0] (
09:06:58 Quit perrikwp_ (Ping timeout: 252 seconds)
09:08:54 Quit scorche (Disconnected by services)
09:08:59 Join scorche` [0] (~scorche@rockbox/administrator/scorche)
09:10:33 Join petur [0] (~petur@rockbox/developer/petur)
09:19:11 Quit bertrik (Ping timeout: 245 seconds)
09:25:18 Quit scorche` (Ping timeout: 252 seconds)
09:26:04 Join scorche [0] (~scorche@rockbox/administrator/scorche)
09:27:01 Join pamaury [0] (~quassel@
09:27:01 Quit pamaury (Changing host)
09:27:01 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:28:07 Join wodz [0] (
09:29:14wodzpamaury: atjboottool doesn't compile. In line 297 it should have struct ptr_bundle_t *ptrs and it misses MIN() declaration.
09:29:37pamauryyeah, I fixed that but didn't commit
09:31:14wodzheh, usage refers to rkboottool :-)
09:31:57pamauryhaha, I'll fix that later, that's not too important :)
09:32:29pamauryI'm currently stuck on the reverse engineer of a routine I really don't understand, but otherwise I'm pretty close to the end I think
09:32:39pamaury(let aside debugging !)
09:33:11wodzalso rather minor thing but annoying is that this tool leave console in yellow color.
09:35:20pamauryah, good remark, I'll fix that too
09:36:54 Join petur2 [0] (
09:36:55 Quit petur (Ping timeout: 252 seconds)
09:38:03 Quit mgottschlag (Read error: Connection reset by peer)
09:41:57 Join mortalis [0] (
09:45:15wodzmortalis: any news with freq scaling problems?
09:46:41 Join einhirn [0] (
09:46:42***Saving seen data "./dancer.seen"
09:50:06 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
09:52:47mortalisi figured out that with nop at the begging of commit_discard_idcache or in any function with lower address i couldn't reproduce data abort. But i don't know how it could help us, it's very weird.
09:52:53mortaliswodz: ^
09:54:17 Join ack [0] (
09:54:58wodzmortalis: what happens if you insert two nops? It data aborts again?
09:55:26pamaurylooks like an alignement problem or cache + dma alignement problem
09:57:20wodzin that case two nops should trigger crash as well, no?
09:58:26mortalisalso address in sp after data abort always points to CODEC_MAGIC or PLUGIN_MAGIC. It depends if I start playback or some plugin,
10:00:01 Quit pamaury (Ping timeout: 252 seconds)
10:02:19mortalisI guess it's so because it somehow correlated with plugin and codec loading.
10:02:30 Nick petur2 is now known as petur (
10:02:47 Quit petur (Changing host)
10:02:47 Join petur [0] (~petur@rockbox/developer/petur)
10:03:11wodzmortalis: which makes commit_discard_idcache() suspicious
10:04:02wodztry with disabled caches if this helps we will narrow the area to investigate
10:12:08 Join Zagor [0] (
10:12:08 Quit Zagor (Changing host)
10:12:08 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
10:34:46mortaliswodz: seems disabling cache helped
10:35:06 Join Keripo [0] (
10:35:21wodzok, so I strongly suspect commit_discard_idcache()
10:35:47mortalisit's pretty simple function
10:35:56mortalisand looks good
10:37:52 Quit mgottschlag (Ping timeout: 260 seconds)
10:39:28wodzcommit_discard_idcache() is called when loading plugins and codecs. If disabling caches helps it is strong indication that something is fishy with cache handling.
10:53:33kugelsince disabling caches makes things slower it could also point to a timing issue (however i dont think that's the case here)
10:57:01kugelbefore plugins are executed the icache is discarded since there's new code in ram. I guess the same needs to be done after relocating since the code in ram changed again. I haven't looked at the patch but I guess you already do this?
10:58:05kugelthe rb->commit_discard_idcache() in plugin_crt0.c shouldnt be needed iiuc
10:58:37 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:58:58pamauryI'm quite confident that the cache handling is correct for plugins and codecs
10:59:40kugelin fact it may be wrong. the commit_dcache() call is to make sure other cores see the zeroing of bss memory; discarding the cache before would do the wrong thing (not all bss might be zero)
11:00:29kugelpamaury: master or with the bFLT patch?
11:00:55kugelhm i thought they talk about the bFLT patch
11:01:14kugelyea, master should be correct
11:01:15pamauryI mean that a good reference to follow
11:01:16 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
11:02:45wodzkugel: We are talking mysterious about data aborts in master on rk27xx targets after enabling freq scaling.
11:02:56 Quit pedro_angelo (Remote host closed the connection)
11:02:56kugeloh, sorry then :)
11:03:01wodzs/talking/talking about/
11:03:37pamauryI have similar problems on imx233, frequency scaling is unable
11:10:44wodzBTW. The reason for data aborts/freezes with current bFLT patch is rather simple. Branching on arm is restricted to 32MB region. If iram is further in memmap branching needs to go through veneer function.
11:11:36wodzCurrent approach neglected this limitation which leads to unpredictable behavior.
11:21:16 Quit Galois (Ping timeout: 260 seconds)
11:21:26 Quit Rondom_ (Remote host closed the connection)
11:21:38 Quit freqmod (Quit: No Ping reply in 180 seconds.)
11:21:44 Quit Elfish (Ping timeout: 260 seconds)
11:22:10 Quit LinusN (Read error: Connection reset by peer)
11:22:16 Quit megal0maniac (Read error: Connection reset by peer)
11:22:16 Join VSNhSI`K [0] (
11:22:17 Join freqmod_abakus [0] (~quassel@2001:700:300:2120:5652:ff:fe70:1e36)
11:22:17 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
11:23:11 Join Elfish [0] (amba@2a01:4f8:100:90a1:abc:abc:abc:abc)
11:28:27 Quit VSNhSI`K (Remote host closed the connection)
11:28:38 Join LinusN [0] (
11:30:07mortaliswodz: 8 nops and again data aborts.
11:30:58wodzisn't it exactly cacheline size?
11:31:07mortalisyes iirc
11:32:29wodzDoes it matter if nops are at the begining or at the end of commit_discard_idcache()?
11:33:46 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
11:34:07mortalisat the and of commit_discard_idcache data aborts happend even without nops.
11:34:31wodzok, so this is THIS function.
11:35:42wodzmaybe actual write to cache controller must be at some particular alignment?
11:36:48kugelwasnt this the case on some PP502x also?
11:37:00kugelor PP5002?
11:37:30wodzkugel: yes in one of this you have to issue sleep command at particular alignment
11:37:45wodzotherwise strange things happen (tm)
11:46:43***Saving seen data "./dancer.seen"
11:51:07 Quit ack (Quit: Coyote finally caught me)
11:59:41 Quit GodEater (Quit: leaving)
12:00:12 Join GodEater [0] (
12:00:13 Quit GodEater (Changing host)
12:00:13 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
12:02:24 Join [Saint_] [0] (~saint@rockbox/user/saint)
12:03:34 Quit [Saint] (Ping timeout: 272 seconds)
12:11:57 Nick [Saint_] is now known as [Saint] (~saint@rockbox/user/saint)
12:12:09 Quit [Saint] (Quit: Quit)
12:12:30 Join [Saint] [0] (~saint@rockbox/user/saint)
12:24:04 Join lorenzo92 [0] (
12:25:09lorenzo92I'm trying to enable usb support (using kernel tools) in samsung ypr0, I want to use functions like usb_enable, disable and detect but which config to enable??
12:25:23lorenzo92I cannot find any documentation :(
12:25:39lorenzo92what is USE_ROCKBOX_USB for example?
12:26:37gevaerts*don't* set USE_ROCKBOX_USB or HAVE_USBSTACK
12:27:16lorenzo92uhm okay I'll try, because without anything gives an error, wait..
12:27:50 Join megal0maniac [0] (~quassel@
12:28:30lorenzo92uhm: /home/lorenzo/YP-R0_DEV/rockbox/rockboxUSB/firmware/usb.c: In function 'usb_slave_mode':
12:28:32lorenzo92/home/lorenzo/YP-R0_DEV/rockbox/rockboxUSB/firmware/usb.c:348: error: expected expression before ';' token
12:28:42wodzlorenzo92: take some target with hardware usb<->ide bridge as reference. For example mpio's or archoses
12:31:40gevaertsYou'll need some APP ifdefs there I guess
12:32:27gevaertsstorage_init() might or might not be applicable to app builds
12:43:10 Quit mgottschlag (Ping timeout: 246 seconds)
12:44:02 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
12:50:48lorenzo92yep I'll guessed that...need to find out where and what :)
12:52:16 Quit mgottschlag (Ping timeout: 246 seconds)
12:53:51lorenzo92okay thanks, I have probably find out what to isolate ;)
12:54:43 Join megal0maniac_ [0] (~quassel@
12:56:29 Quit megal0maniac (Ping timeout: 252 seconds)
12:58:51 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
13:02:55 Join kakonema [0] (6d5d0303@gateway/web/freenode/ip.
13:08:39 Quit kakonema (Quit: Page closed)
13:09:32 Join kakonema [0] (
13:09:45 Quit kakonema (Client Quit)
13:09:53 Join kakonema [0] (
13:12:07 Quit kakonema (Client Quit)
13:18:37lorenzo92usb build compiles fine, but nothing happens, do I need to use USB_STATUS_BY_EVENT or so? who is actually calling usb_detect?
13:23:10gevaertsWell, IIRC either you define USB_STATUS_BY_EVENT and then *you* call usb_detect() and then usb_status_event() whenever something changes, or you *don't* define USB_STATUS_BY_EVENT and you get a tick task that calls usb_detect() periodically
13:24:06lorenzo92I'm for the second, but then something is missing, I need to understand better usb code :)
13:25:39 Quit megal0maniac_ (Quit: - Chat comfortably. Anywhere.)
13:27:21 Join megal0maniac [0] (~quassel@
13:28:54 Join einhirn [0] (
13:29:06 Quit pamaury (Ping timeout: 240 seconds)
13:30:47 Join pamaury [0] (
13:30:47 Quit pamaury (Changing host)
13:30:47 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
13:31:49 Quit mgottschlag (Ping timeout: 246 seconds)
13:37:11 Quit thegeek_ (Ping timeout: 252 seconds)
13:44:20lorenzo92okay I've added the USB_FULL_INIT (that initializes and runs the usb thread...)
13:46:47***Saving seen data "./dancer.seen"
13:50:08 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
13:52:04lorenzo92quick question: were are screendumps stored?
13:53:42gevaertsHOME_DIR, which seems to be /mnt/media0 for you
13:53:48gevaerts(if I read things correctly)
13:54:32lorenzo92yes thanks :)
13:54:50lorenzo92I'm quite done, still to see why usb isn't recognized, perhaps I did something wrong hehe
14:03:47lorenzo92got it!! ;) uhm possible that screens are going into the root folder? in this case in the R0 is readonly...
14:03:55lorenzo92I'll do a little check
14:06:31 Nick freqmod_abakus is now known as freqmod (~quassel@2001:700:300:2120:5652:ff:fe70:1e36)
14:09:00 Quit megal0maniac (Read error: Connection reset by peer)
14:09:07 Join megal0maniac [0] (~quassel@
14:11:20lorenzo92indeed, I will need to do a little change in the code, anyways screendump crashes on R0 will see that later why ;)
14:14:18 Quit wodz (Quit: Leaving)
14:14:44 Join amayer_ [0] (
14:22:35 Quit lorenzo92 (Quit: ChatZilla [Firefox 15.0.1/20120907231657])
14:22:56 Quit [Saint] (Read error: Connection reset by peer)
14:24:51 Join [Saint] [0] (~saint@rockbox/user/saint)
14:46:31 Nick n17ikh_ is now known as n17ikh (~n17ikh@unaffiliated/n17ikh)
14:48:15 Quit Keripo (Quit: Leaving.)
14:54:24 Join lebellium [0] (
15:02:55 Part amayer_
15:15:08 Join amayer_ [0] (
15:34:10 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
15:37:07 Join megal0maniac_ [0] (~quassel@
15:40:58 Quit megal0maniac (Ping timeout: 255 seconds)
15:46:48***Saving seen data "./dancer.seen"
15:52:01 Quit mortalis (Quit: Leaving)
16:05:45amayer_does rockbox have a feature to log all usb traffic?
16:05:47amayer_i googled "log usb" but couldnt fine anything relevent
16:07:03amayer_gevaerts: thank you
16:08:19 Quit sinthete1 (Ping timeout: 246 seconds)
16:10:06 Join Strife89 [0] (
16:11:54 Quit [Saint] (Remote host closed the connection)
16:13:42 Join [Saint] [0] (~saint@rockbox/user/saint)
16:14:21 Part LinusN
16:19:09 Join einhirn [0] (
16:31:50saratogaamayer_: using the ERRORF/WARNF/NOTEF macros you could log USB errors or other changes
16:32:04saratogayou'd have to know where to put the debug statements though :)
16:32:56amayer_im looking for the easiest way to reverse engineer the apple protocal with stereo connections
16:33:24amayer_so if i can dump the communication between the stereo and my ipod i figured that would be a good start
16:33:54saratogatheres actually detailed specifications for the apple protocol floating around
16:34:21amayer_really? any on the wiki?
16:34:22amayer_or were you talking about outside sources?
16:34:33saratogabut IIRC it uses public key encryption to verify USB audio connections, so that might be tough to implement unless you can figure out one of their keys
16:35:27saratogacheck google, there should be a datasheet
16:43:27amayer_thank you very much. will do
16:44:20 Join Epicanis [0] (
16:50:44 Quit megal0maniac_ (Read error: Connection reset by peer)
16:51:33 Join megal0maniac [0] (~quassel@
17:00:10 Quit kevku (Quit: KVIrc 4.2.0 Equilibrium
17:02:17 Quit Zagor (Quit: Clint excited)
17:04:29 Join pretty_function [0] (~sigBART@
17:11:15 Quit mgottschlag (Ping timeout: 246 seconds)
17:12:48 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
17:17:48 Join efyx [0] (~efyx@
17:28:45 Part eckoit
17:32:48 Join tchan1 [0] (
17:33:22 Quit Strife89 (Quit: Vamoose)
17:35:34 Quit tchan (Ping timeout: 268 seconds)
17:39:40 Join thegeek [0] (
17:46:51***Saving seen data "./dancer.seen"
17:50:06 Quit thegeek (Quit: ( :: NoNameScript 4.2 :: ))
17:52:44 Quit [7] (Disconnected by services)
17:52:52 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
17:53:28 Join LambdaCalculus37 [0] (~LambdaCal@rockbox/staff/LambdaCalculus37)
17:54:07LambdaCalculus37TheSeven: Ping
17:55:11 Join eckoit [0] (~ryan@
17:56:18*LambdaCalculus37 has another iPod Classic in his possession
17:56:36LambdaCalculus37USB issues still about with OS X.
17:59:19 Quit pretty_function (Remote host closed the connection)
18:04:56TheSevenLambdaCalculus37: USB issues with pretty much everything recently, eh?
18:05:08LambdaCalculus37TheSeven: Yeah, pretty much. :)
18:06:54LambdaCalculus37TheSeven: I installed emCORE r859 and installed via Linux with the instructions here:
18:07:40LambdaCalculus37That went smoothly enough; the drive was wiped, and I installed build rev 1c975ee-120916.
18:08:19LambdaCalculus37But one question: I remember that the drive is formatted as a "superfloppy" via emCORE; is this still the case?
18:10:05TheSevenusually yes
18:10:19TheSevenbut apparently it doesn't always get formatted at all
18:10:27TheSevenI still don't get why
18:11:16user890104TheSeven: i can confirm that, i still have ipod_control and the other itunes-related folders on my classic (without re-uploading them back after installation)
18:11:19LambdaCalculus37I have to use a Windows or Linux box to access the filesystem. OS X just herp-derps and refuses to mount the volume.
18:11:38TheSevenoh, funny, especially if it wasn't even formatted
18:11:41user890104LambdaCalculus37: can you try emCOREFS?
18:11:42TheSevenwhat's the device name in linux?
18:11:50LambdaCalculus37And I no longer have the ipod_control folder on the volume. emCORE did format the volume.
18:11:50user890104it should work on OSX
18:11:54TheSeven/dev/sdX or /dev/sdX1 or /dev/sdX2?
18:12:06TheSevenah ok
18:12:28 Join WalkGood [0] (~4@unaffiliated/walkgood)
18:12:42LambdaCalculus37TheSeven: In OS X, that would be /dev/diskXsY (X is disk number, Y is volume)
18:12:53LambdaCalculus37i.e. My Macbook HDD is /dev/disk0.
18:13:05TheSevenyeah, I meant the successfully mounted one on linux
18:13:34LambdaCalculus37TheSeven: Let me get it.
18:13:49TheSevenprobably just /dev/sdX then, which explains the OSX trouble
18:14:02LambdaCalculus37TheSeven: It mounts as /dev/sdb
18:14:03TheSevenone thing that you might try is formatting it on windows
18:14:14TheSeventhat should work on all OSes afterwards
18:14:27 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
18:14:46LambdaCalculus37TheSeven: I'll try that then. I didn't copy anything new to it, thankfully, so I can just clone my Gigabeat's /Music folder straight back into it afterwards. :)
18:14:52TheSevenor maybe formatting on osx if that allows to format it with fat and doesn't insist on hfs+
18:15:20LambdaCalculus37TheSeven: Disk Utility will allow me to do so… that is, IF the disk actually shows up.
18:17:20*TheSeven wonders if this is caused (indirectly?) by the superfloppy format, or something totally different
18:17:47LambdaCalculus37TheSeven: More thorough output from my syslog:
18:23:36 Quit pamaury (Ping timeout: 260 seconds)
18:23:55TheSeventhis log isn't particularly helpful because it doesn't say what it wad trying to do when it got the timeouts
18:24:39TheSevenbut it could possibly make more sense in combination with a usb trace
18:24:42 Join thegeek [0] (
18:25:00 Quit [Saint] (Ping timeout: 260 seconds)
18:25:25LambdaCalculus37TheSeven: I have noticed that the iPod tends to power down after about a minute or two on USB.
18:27:39amayer_you mean it shuts off totally?
18:30:21 Quit mgottschlag (Ping timeout: 246 seconds)
18:30:37LambdaCalculus37Yes, it does.
18:30:44LambdaCalculus37On my Mac, anyway.
18:32:08amayer_oh. i use my ipod classic on linux almost every day. ive never had a problem with power down.
18:32:10amayer_i do have problems with it going into usb mode when connected to charger(wall outlet) tho
18:32:53 Join [Saint] [0] (~saint@rockbox/user/saint)
18:34:02 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
18:42:36 Quit mgottschlag (Ping timeout: 246 seconds)
18:47:00LambdaCalculus37amayer_: I have that issue as well (going into USB mode on a wall charger), which I remedy by holding MENU as I plug it in.
18:48:12amayer_yeah i do the same thing.
18:48:14amayer_its just annoying when i forget
18:48:54gevaertsThen please fix it :)
18:49:58gevaertsAll you need to do is enable USB_DETECT_BY_REQUEST, and if needed figure out what's broken about it
18:55:12amayer_is that a setting or something i need to recompile for?
18:57:16 Quit nosa-j (Read error: Connection reset by peer)
18:57:38 Join kevku [0] (
18:58:59 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
18:59:32 Join nosa-j [0] (~m00k@
19:09:13 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
19:14:46 Quit WalkGood (Quit: ♪ ♫ ♪ ♫ ♪ ♫ ♪)
19:16:30 Join TheLemonMan [0] (~LemonBoy@unaffiliated/thelemonman)
19:46:53***Saving seen data "./dancer.seen"
19:56:42gevaertsThat's a setting, in config.h
19:57:27TheSevenbut one that requires recompiling
19:58:33gevaertsRight. That wasn't worded very well
20:00:53amayer_gevaerts: sorry i ment a setting in the menu(post compile) or pre compile C constant
20:01:16gevaertsIt's a preprocessor define :)
20:03:31amayer_what does that do after you compile with that option. i cant find anything on the wiki about it
20:04:08gevaertsIt changes USB detection, so an actual host os needed to go to disk mode
20:04:21TheSevenit basically tries to judge on whether the thing it's connected to is a charger or a pc by looking for actual USB traffic
20:04:21gevaertsInstead of relying on various heuristics than can fail
20:10:21amayer_so does it give you some kind of id or "connection state" that would let you know it is a charger vs a pc?
20:10:44gevaertsWe wait for actual enumeration
20:11:05TheSeveni.e. assume it's a charger until it actually starts communicating with us, right?
20:11:35TheSevenhm... not very specs compliant :P
20:11:45amayer_*scratches head*
20:11:47amayer_so the other way we just always assume its a pc?
20:12:08gevaertsThe other way we assume that if the controller reports a bus reset, we're on a host
20:12:26amayer_oh ok. so some chargers will cause a bus reset
20:12:36gevaertsUnfortunately some controllers or some chargers seem to cause that too often
20:12:46 Join Wardo [0] (
20:12:55TheSevengevaerts: usb-s3c6400x triggers a bus reset IRQ while powering up the core
20:13:09TheSevenwe might want to check for an "enumeration done" IRQ instead
20:13:25TheSevenwhich occurs basically when the host and OTG have negotiated on a communication speed
20:13:45gevaertsThat would be better than bus reset, yes.
20:14:05TheSevenan probably not much different than waiting for an actual get device descriptor request
20:14:06amayer_ive worked with C before
20:14:08amayer_but ive never done anything this low level before.
20:14:18Tornewe dont assume its a charger tgat way
20:14:32Torneat least the targets i fixes dont and there is a framework for it
20:14:34gevaertsNote that that's *host* detection. On devices where Torne has touched the code, we don't assume a real charger too easily if we don't detect a host
20:14:39Tornesee pp502x
20:15:01TheSevenand what do we do if we have neither a charger or host? charge with 100mA?
20:15:24Torneyes, that is the ibtention
20:15:38Tornepp502x does it 'right'
20:15:38 Quit GodEater (Ping timeout: 252 seconds)
20:15:49Tornewell as far as i have gotten around to.
20:15:55TheSevenwhich devices do actual USB line bitbanging to check for charging spec compliant chargers and pull 1000mA? :)
20:16:06Tornenot implemented anywhere.
20:16:12Torneframework is there.
20:16:21Torneyou jus need to write the detection code
20:16:23TheSevenat least on the classic we even have ADCs on D+/D-
20:16:30Tornei was working on it on ipodvideo
20:16:54Torneanyway i am still on vacation
20:16:57Tornebug me later :)
20:18:54Torneif someone works out the detction for some target and sends me a patch i will hook it up to my charging framrwork
20:19:06Torneor review you doing that yourself. either way
20:21:03 Join sinthetek [0] (
20:21:03 Quit sinthetek (Changing host)
20:21:03 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
20:23:52gevaertsamayer_ (and LambdaCalculus37!): you'd start from
20:25:13kugellorenzo92 (logs): If screendump crashes it could be a stack overflow; the screendump code uses on-stack line buffers on the callers stack (normally the usb thread)
20:27:31 Quit mikroflops (Ping timeout: 248 seconds)
20:28:38 Join mikroflops [0] (
20:31:48 Join pamaury [0] (~quassel@
20:31:48 Quit pamaury (Changing host)
20:31:48 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:35:19 Join GodEater [0] (
20:35:19 Quit GodEater (Changing host)
20:35:19 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
20:41:19 Join liar [0] (
20:41:41 Quit Wardo (Quit: Blarglarg)
20:44:46 Join lorenzo92 [0] (
20:45:23lorenzo92kugel: uhm actually I got the screenshot in the folder, perhaps there is something further ... I'll investivage now
20:52:54kugellorenzo92: you can die on a stack overflow and still have a screendump (it may even look right)
20:53:19lorenzo92in fact the dump is 100% okay
20:53:29lorenzo92but it does the same in "usb" mode
20:53:37lorenzo92so i don't think that's really the problem
20:54:29kugelcan I see your current code? I'm curious
20:55:04lorenzo92well I have been away this afternoon :D I'll clean it, and such then yes, of course ;)
20:55:21lorenzo92we use the linux gadgets, like the safe mode
20:56:35kugelno need to clean it up, I just want to see hte approach, I dont want to test or even compile it
20:56:45lorenzo92ah well for the moment we use a system call to call a script, later I will include everything in the code (system calls)
20:56:50lorenzo92okay, wait a second ;)
20:57:31 Join Horscht [0] (~Horscht@xbmc/user/horscht)
20:59:00 Join pretty_function [0] (~sigBART@
21:02:38 Quit Horscht (Quit: Verlassend)
21:03:18 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
21:04:29lorenzo92kugel: pff I still have to get used with git...I added the new file with -add and now why doesn't it go into the patch as well?
21:05:54 Quit TheLemonMan (Quit: WeeChat 0.3.8)
21:13:08kugellorenzo92: try git diff HEAD
21:14:09lorenzo92yeah that was ^^ thanks ;)
21:14:14kugelgit diff by default shows the diff to the index (i.e. unstaged changes). git add adds stuff to the index (stages) as well
21:14:37kugelwith HEAD you specify to compare against the most recent commit, not the index
21:15:08kugeli know, the HEAD vs index thing is often confusing if you dont know about it
21:16:21lorenzo92kugel: :)
21:19:32lorenzo92could we create a new variable like: SCRDUMP_DIR
21:19:46lorenzo92because in R0 the root / isn't RW
21:20:00kugelwhat does usb_on/usb_off do?
21:20:26kugeli think we can use HOME_DIR for all targets, it should be / for native targets
21:20:37 Join Wardo [0] (
21:21:11 Quit pretty_function (Ping timeout: 246 seconds)
21:21:39lorenzo92kugel: okay good for the path, usb_on and off ... wait :p
21:21:53kugelI had to fix create_numbered_filename() to accept "/" instead of "" for files in the root for a similar problem
21:22:30lorenzo92hehe so I leave home_dir...
21:23:54 Quit benedikt93 (Quit: Bye ;))
21:24:47lorenzo92but I'd prefer doing all this stuff in rockbox
21:25:06lorenzo92vanniX discovered that this gadget works also with rockbox opened
21:25:22lorenzo92so it should be completely safe...
21:26:05lorenzo92note: g_serial is because we added the serial through the usb in another firmware ;)
21:26:33 Join pretty_function [0] (~sigBART@
21:27:04kugelwell, I expected it would work with rockbox too
21:27:48lorenzo92yep also without umount the disks...
21:28:33kugelI guess the kernel unmounts internally
21:29:27lorenzo92uhm ... anyways need to figure out what's the current issue (freeze after usb has been called, either screendump or disk mode)
21:29:50 Quit megal0maniac (Remote host closed the connection)
21:34:18lorenzo92uhm do you think I should try to enlarge the stack for test?
21:40:56lorenzo92okay it doesn't help, I did 4 times this stack: the same
21:44:31lorenzo92don't understand...perhaps something related to my usb_detect code?
21:46:56***Saving seen data "./dancer.seen"
21:55:08lorenzo92I'll check with the serial...
21:56:33lorenzo92pff I can't xD I use the usb for the serial :p
22:00:37 Quit pretty_function (Quit: Leaving...)
22:01:59 Join Horscht [0] (
22:01:59 Quit Horscht (Changing host)
22:01:59 Join Horscht [0] (~Horscht@xbmc/user/horscht)
22:11:18lorenzo92kugel: Segmentation fault!
22:14:26 Quit liar (Ping timeout: 252 seconds)
22:16:30lorenzo92kugel: errata "works" normally, crashes with screendump
22:16:48 Join liar [0] (
22:43:03 Join megal0maniac [0] (
22:43:59 Quit lorenzo92 (Quit: ChatZilla [Firefox 15.0.1/20120907231657])
22:47:57 Quit LambdaCalculus37 (Read error: Connection reset by peer)
22:48:10 Join LambdaCalculus37 [0] (~LambdaCal@rockbox/staff/LambdaCalculus37)
22:50:19 Quit LambdaCalculus37 (Read error: Connection reset by peer)
22:50:43 Join LambdaCalculus37 [0] (~LambdaCal@rockbox/staff/LambdaCalculus37)
22:51:57 Nick zz_TheSphinX^ is now known as TheSphinX^ (
22:52:28 Quit LambdaCalculus37 (Read error: Connection reset by peer)
22:52:46 Join LambdaCalculus37 [0] (~LambdaCal@rockbox/staff/LambdaCalculus37)
22:53:43 Join lebellium_ [0] (
22:55:30 Quit lebellium (Ping timeout: 252 seconds)
22:55:37 Nick lebellium_ is now known as lebellium (
23:03:30 Join Galois [0] (
23:11:21 Quit liar (Quit: huiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii)
23:17:13 Quit Epicanis (Quit: got to go, possibly back later...)
23:19:33 Part amayer_
23:20:26 Quit Horscht (Quit: Verlassend)
23:20:49 Join lorenzo92 [0] (
23:21:20lorenzo92kugel: funny that after inserting some printfs, screendump works correctly now :o
23:22:29 Quit XavierGr (Ping timeout: 272 seconds)
23:26:34 Quit lorenzo92 (Quit: ChatZilla [Firefox 15.0.1/20120907231657])
23:27:33 Quit kevku (Ping timeout: 272 seconds)
23:31:37 Quit Rower85 (Read error: Connection reset by peer)
23:41:12 Quit ender` (Quit: The process takes just over six seconds, and would sound like a coffee machine passing an aluminum baseball bat if there were any atmosphere to carry the sound.-- Howard Tyler: Schlock Mercenary)
23:45:40 Join webguest15 [0] (
23:46:11 Quit webguest15 (Client Quit)
23:47:00***Saving seen data "./dancer.seen"
23:53:11 Join trux_ [0] (
23:57:54 Join amayer [0] (
23:58:27 Quit megal0maniac (Remote host closed the connection)

Previous day | Next day