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

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

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

#rockbox log for 2012-05-26

00:00:06gevaertsShould be easy, but I want to add a comment clarifying that expression first
00:01:45saratogaalso, remove the ifp700 case
00:02:09 Quit TheSeven (Disconnected by services)
00:02:16saratogathat platform hasn't been maintained in many years, and that define almost certainly won't work for it
00:02:17 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
00:02:37gevaertsThat's the point I think. It doesn't have the necessary code for callbacks.
00:02:54gevaertsI agree though, unmaintained and unfinished ports shouldn't make our code hard to read
00:03:47gevaertsSo it needs to be enabled for hosted and for native non-bootloader. I think that basically means non-bootloader
00:04:48saratogai doubt the ifp builds anyway, so if someone ever picks up that port they can fix that as well
00:04:57*gevaerts nods
00:05:25saratogaalthough given that it only has 1MB of RAM (IIRC), they'll probably need to use a much older version of rockbox
00:06:10gevaertsI doubt if anything will ever happen there
00:10:08CIA-79Commit 3f72ba0 in rockbox by Frank Gevaerts: Fix broken logic for enabling USING_STORAGE_CALLBACK.
00:13:02CIA-793f72ba0 build result: All green
00:13:54saratogai'll close the bug report
00:14:30*gevaerts adds a note to the bug report explaining what happened
00:14:54gevaertsLong conditions that aren't purely lists of && or || are fragile :\
00:18:45 Join megal0maniac [0] (
00:18:53bluebrotherindenting braced conditions can help :)
00:19:24bluebrotherhmpf, why is the size field in voice files reused? Makes it harder to eliminate it :(
00:19:43gevaertsIt can, but simplifying things to !BOOTLOADER is even better :)
00:20:09megal0maniacAnyone seen this?
00:20:32megal0maniacNot sure if it's a repurposed mp3 player, or just in the shell of one
00:20:35gevaertsmegal0maniac: yes. Last weekend :)
00:20:37bluebrothermegal0maniac: that says "retired"
00:20:49bluebrotherand we saw a v2 last weekend (I have one :)
00:21:03bluebrotherthese days there's even a "DSO quad" available
00:21:05bertriksaratoga, yes, I've seen slow volume change on the zen xfi3, and spinup on volume change on a samsung yh-xxx from gevaerts
00:21:18megal0maniacYeah. But the V2 isn't relevant to this because it doesn't look like a player :)
00:21:18bluebrother(though that is a bit pricey imo)
00:21:30megal0maniacIs it an mp3 player at heart?
00:21:47megal0maniacSince technically speaking, it should be possible to convert
00:22:11bluebrotherAFAIK they originally started with some mp3 player design
00:23:08megal0maniacHeard so too. Curious as to how much was actually changed hardware wise
00:25:02bluebrotherthough I just heard about that and have no souce, so I oculd be wrong
00:25:02megal0maniacI'm sorry, because I know this is off topic, but is anyone here clued up on AVR processors? (Atmega in particular)
00:25:02bluebrotherfor the v2 the schematics are available
00:25:02 Join Syconaut^ [0] (
00:25:02bluebrothermegal0maniac: guess this would be a topic for #rockbox-community then :)
00:25:21bluebrotherbut I guess quite a few people do. I did some AVR stuff years ago
00:25:43 Quit Zambezi (Ping timeout: 260 seconds)
00:26:12 Join Zambezi [0] (
00:26:35megal0maniacMoving my question to #rockbox-community. Ta :)
00:26:55 Quit stoffel (Remote host closed the connection)
00:27:24 Quit guymann (Ping timeout: 252 seconds)
00:28:04 Join guymann [0] (
00:28:15 Quit Syconaut (Ping timeout: 252 seconds)
00:29:29bertrikah, volume change is much more responsive now
00:31:32saratogasince that Nano thing is a cortex M3, I doubt it was an MP3 player
00:32:15saratogai don't even think you can have DRAM on those, which makes decoding audio kind of tough (or at least very expensive if you want enough SRAM to do it)
00:35:12bertrik.describe 6b8330d2
00:35:13fs-bluebotEnable USING_STORAGE_CALLBACK on hosted targets. Fixes config.cfg not written properly. by Thomas Martitz (from Sun, 13 May 2012 16:56:38 +0000)
01:00:43 Quit nosa-j (Excess Flood)
01:01:08 Join Synergist [0] (~synfn@unaffiliated/synergist)
01:03:18 Quit soap (Quit: soap)
01:03:23 Join nosa-j [0] (
01:04:51 Join soap [0] (
01:04:51 Quit soap (Changing host)
01:04:51 Join soap [0] (~soap@rockbox/staff/soap)
01:06:55 Join bieber [0] (~quassel@
01:10:05 Join bitcraft [0] (
01:10:41 Quit bitcraft (Remote host closed the connection)
01:12:12 Quit liar (Remote host closed the connection)
01:13:24 Quit megal0maniac (Quit: Konversation terminated!)
01:20:27 Join remlap [0] (
01:22:50 Join bitcraft [0] (
01:25:10 Quit bertrik (Ping timeout: 246 seconds)
01:26:39 Quit Rower85 (Quit: Hmmm...)
01:46:11***Saving seen data "./dancer.seen"
02:00:19 Join eyfour [0] (
02:04:20 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
02:05:08 Quit mystica555 (Quit: Ekkusu Chatto)
02:05:46 Join mystica555 [0] (
02:05:53 Quit mystica555 (Max SendQ exceeded)
02:07:48 Part eyfour ("WeeChat 0.3.7")
02:18:36 Quit ender` (Quit: Documentation is like sex: when it's good, it's very good, and when it's bad it's still better than nothing.)
02:36:44 Nick Cthulhux` is now known as Cthulhux (cthulhux@piratenpartei/ni/tux)
02:53:22 Quit lebellium (Quit: ChatZilla [Firefox 13.0/20120516113045])
02:59:55 Join Kaiscene [0] (
02:59:56 Quit Kaiscene (Changing host)
02:59:56 Join Kaiscene [0] (~Kaiscene@unaffiliated/kaiscene)
03:05:52 Quit hillshum (Ping timeout: 250 seconds)
03:06:28 Quit LambdaCalculus37 (Quit: Fwump)
03:27:37 Quit kadoban (Ping timeout: 244 seconds)
03:40:36 Quit jfc (Ping timeout: 240 seconds)
03:46:13***Saving seen data "./dancer.seen"
03:53:53 Join jfc [0] (
03:55:05 Quit XavierGr ()
04:00:41 Quit Horscht (Quit: Verlassend)
04:12:41 Join Sundiver_ [0] (
04:16:20 Quit Sundiver (Ping timeout: 248 seconds)
04:21:50 Quit jfc (Ping timeout: 252 seconds)
04:25:02 Join jfc [0] (
04:31:10 Join TheSphin_ [0] (
04:34:43 Quit zz_TheSphinX^ (Ping timeout: 245 seconds)
04:40:22 Quit [7] (Disconnected by services)
04:40:29 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
04:44:20 Quit amiconn (Disconnected by services)
04:44:22 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:44:26 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:46:44 Quit pixelma (Disconnected by services)
04:46:46 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:46:48 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
05:34:18 Join kevku [0] (
05:41:47 Join mystica555 [0] (
05:42:32 Quit mystica555 (Client Quit)
05:46:11 Join mystica555 [0] (
05:46:16***Saving seen data "./dancer.seen"
05:46:39 Quit mystica555 (Remote host closed the connection)
05:51:32 Join mystica555 [0] (
05:59:31 Quit nosa-j (Read error: Connection reset by peer)
06:01:56 Join nosa-j [0] (
06:02:26 Quit nosa-j (Excess Flood)
06:03:26 Join nosa-j [0] (
06:31:25 Join icarusfactor [0] (
06:32:07 Quit factor (Ping timeout: 240 seconds)
06:33:36 Join kadoban [0] (
06:43:40 Quit anewuser ()
06:45:04 Quit prof_wolfff (Ping timeout: 246 seconds)
07:26:41 Nick icarusfactor is now known as factor (
07:34:40 Quit kevku (Quit: KVIrc 4.0.4 Insomnia
07:37:04 Quit factor (Read error: Connection reset by peer)
07:46:20***Saving seen data "./dancer.seen"
07:54:55 Join factor [0] (
08:12:19 Join stoffel [0] (
08:23:22 Quit BHSPitMonkey (Remote host closed the connection)
09:00:24 Quit bluebrother (Disconnected by services)
09:00:31 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
09:03:40 Quit fs-bluebot (Ping timeout: 246 seconds)
09:04:47 Join fs-bluebot [0] (
09:07:34 Join Rower85 [0] (
09:08:26 Quit nosa-j (Read error: Connection reset by peer)
09:08:31 Join nosa-j [0] (
09:10:29 Join perrikwp_ [0] (
09:12:01 Join rarog [0] (
09:12:35 Join bertrik [0] (
09:12:35 Quit bertrik (Changing host)
09:12:35 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
09:12:47rarogHello, is it possible to direct the output of logf() into a file "out of the box" without patc hes?
09:13:17 Quit perrikwp (Ping timeout: 244 seconds)
09:13:24rarogGoogle only told me about older patches to do this
09:27:29 Quit bitcraft (Remote host closed the connection)
09:46:22***Saving seen data "./dancer.seen"
10:00:24funmanyou can do that from the logf entry in debug menu i think
10:15:07 Quit kadoban (Ping timeout: 240 seconds)
10:19:28pixelmaok, I'll update too. Maybe the storage callback fix also takes care of the never-stopping disk access I saw on the Ondio
10:38:05bertrikpixelma, probably
10:40:22 Join kevku [0] (
10:42:31 Quit stoffel (Ping timeout: 245 seconds)
10:42:58rarogfunman: system -> debug?
10:43:11rarogat least on fuze+ I can't see any settings
10:44:39rarogor is this only if I compile rockbox with some debug flags enabled?
10:54:49funmanyou must enable logf in configure
11:02:35 Nick Totalled is now known as TotalledZzZz (
11:17:15 Quit rarog (Quit: ChatZilla [SeaMonkey 2.9.1/20120501173307])
11:19:04bertriksilly idea, could we print the git hash in the panic backtrace screen? this way it's probably easier to analyse user-provided panic screens
11:19:27Tornewe could, but i'm not sure how much room there is on all targets :)
11:34:08bertrikdo we have some kind of utility to convert a numeric backtrace to source lines?
11:35:15 Join pamaury [0] (
11:35:15 Quit pamaury (Changing host)
11:35:15 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:36:00Tornebertrik: i don't know of one, but if we don't, it's trivial to do
11:36:16Torneaddr2line will pretty much do it for you
11:36:22Tornemaybe a small bash script around it ;)
11:36:28bertrikI vaguely remember seeing a rockbox wiki page about it, but can't find it
11:37:27bertrikI see there's a script in utils/analysis, but it won't run for me
11:38:25Tornearm-eabi-addr2line on the elf binary should do it
11:38:50Torneat least, for addresses in the main binary
11:39:04Tornefind_addr is trying to be clever and handle plugins/codecs
11:39:32Tornei've not used it, have no idea :)
11:40:14bertrikI don't have arm-eabi-addr2line, but I'd expect such a tool to be compiled with our toolchains
11:40:32Tornearm-elf-eabi-addr2line, even
11:40:34Tornewrong prefix
11:41:10Torneaddr2line -e rockbox.elf
11:41:13Tornethen type addresses :)
11:41:22Torneor just pipe a file full of addresses through it
11:41:33bertrikarm-elf-eabi-addr2line -e rockbox.elf 0x300292db gives me "??:0"
11:41:52Tornepossibly that's not in the main binary
11:42:37Tornehm, actually
11:42:44Tornedo we build with -g?
11:43:11Tornei remember being irritated previously :)
11:43:27bertrikI think 0x300xxxxx it somewhere in IRAM or DRAM
11:43:37bertrikfor AMS targets
11:43:54Tornei have no idea. look in the mapfile?
11:45:49bertrikhm, the 0x300..... range is not even mentioned once in the map file, I do remember that we do some clever things with aliasing/remapping stuff on AMS to keep addresses closer (and get smaller code)
11:46:25***Saving seen data "./dancer.seen"
11:46:37Tornethe mapfile is the virtual addresses
11:48:34funmanit is in DRAM but could be in plugin or codec space
11:48:44 Join ender` [0] (
11:49:56funmanbertrik: it is mentioned in fact: it is DRAM_ORIG
11:50:24bertrikI'm looking at,34450.msg198134
11:51:15bertrikfunman, mentioned where? not in the map file
11:51:47funman(you have to know that DRAM_ORIG is around 0x3000... ok)
11:52:19pamaurybertrik: hi, any progress with the tuner ?
11:52:42bertrikpamaury, no, but I haven't really worked on it either
11:53:03pamauryok, can you commit the code or pastebin it if it's not clean ?
11:53:14bertrikok, I'll pastebin it
11:53:46funman% addr2line -e rockbox.elf 0x300292db
11:54:28funmanarm-elf-eabi-addr2line shows libgcc2.c:0
11:54:56funmanbertrik: i added -g to GCCOPTS and rebuilt:
11:55:00funman% arm-elf-eabi-addr2line -e rockbox.elf 0x300292db
11:55:33bertrik-g doesn't increase the target binary, right?
11:55:34funmanit shows the line after the panic (} closing the function)
11:55:49funmanit shouldn't change it at all
11:56:38bertrikpamaury, stfm1000 WIP
11:57:16pamauryI'll try to hack something for the audio later today
11:58:01pamauryin fact it's not that easy because if I understand correctly, the DRI runs at 24MHz (for RDS ?) whereras the audio is at 44.1kHz, so you need to drop many samples probably
11:58:11bertrikpamaury, it's quite crude and doesn't take into account all the weak station special cases etc, but it does find a couple of valid FM stations with automatic FM search
12:00:00bertrikthere is a divider from 24 MHz to set the i2s/dri/sai frequency
12:00:14funmanbertrik: it might be interesting to see when this message first appeared. i would guess c200v2 is too tight for buflib
12:01:58bertrikclip zip has something similar: but is not that memory constrained
12:04:13bertrikfunman, Torne anything against using the -g compiler option? Otherwise I'd say we should add it by default.
12:04:34bertrikperhaps bigger intermediate files, but I don't see that as a big prolem
12:08:14 Join Topy [0] (
12:11:16 Quit T44 (Ping timeout: 246 seconds)
12:14:12 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
12:21:09funmanbertrik: it would be a good idea
12:21:59funmani see -g used in a few Makefiles already
12:25:59 Join TheLemonMan [0] (
12:34:31 Join lebellium_gs2 [0] (
12:38:23 Join wodz [0] (
12:41:36wodzI have been playing a bit with flite TTS. Memory fragmentation is massive but mem pool can be reseted at the begining of each utterance. With the simplest voice I am able to 'speek' standard alice test with ~650kB of heap and dumb allocator (only forward malloc(), no free()). tlsf is much worse in this usage scenario and needs over 800kB of heap.
12:43:48wodzThe real problem seems to be data size needed for engine which are currently compiled into binary itself. We would need to wrap this and load resources on runtime as I don't think plugin buffer is big enough.
13:04:14 Quit wodz (Quit: Leaving)
13:06:22 Join anewuser [0] (~anewuser@
13:06:22 Quit anewuser (Changing host)
13:06:22 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
13:09:40 Join stoffel [0] (
13:23:36 Quit stoffel (Ping timeout: 265 seconds)
13:37:24 Join n1s [0] (~n1s@rockbox/developer/n1s)
13:40:31 Join lebellium [0] (
13:43:57 Quit lebellium_gs2 (Quit: Bye)
13:46:28***Saving seen data "./dancer.seen"
13:55:28 Join prof_wolfff [0] (
14:00:15bertrikpamaury, any luck or problems with the stfm radio code?
14:19:47 Join stoffel [0] (
14:34:01Tornebertrik: we should build with -g, yes. it increases compile time, but possibly not measurably
14:50:08 Quit stoffel (Ping timeout: 244 seconds)
14:54:07 Join lebellium_ [0] (
14:54:15 Quit lebellium_ (Client Quit)
14:55:46 Quit lebellium (Ping timeout: 246 seconds)
14:57:36 Join lebellium [0] (
15:28:28 Join skapazzo [0] (
15:29:04 Quit skapazzo (Client Quit)
15:32:31 Quit lebellium (Ping timeout: 246 seconds)
15:34:24 Join lorenzo92 [0] (~chatzilla@
15:39:21 Join lebellium [0] (
15:46:10lorenzo92kugel: did you find a solution to put the module into the source tree?
15:46:30***Saving seen data "./dancer.seen"
15:46:48 Join Horscht [0] (~Horscht@xbmc/user/horscht)
15:50:19 Join MethoS- [0] (~clemens@
16:26:26 Join stoffel [0] (
16:26:55 Join lebellium_ [0] (
16:28:31 Quit lebellium (Ping timeout: 246 seconds)
16:30:43 Join lebellium [0] (
16:32:01 Quit lebellium_ (Ping timeout: 246 seconds)
16:33:05 Join lebellium_ [0] (
16:35:31 Quit lebellium (Ping timeout: 246 seconds)
16:35:45 Nick lebellium_ is now known as lebellium (
16:41:14 Join kadoban [0] (
16:53:35 Quit stoffel (Remote host closed the connection)
16:57:34 Join stoffel [0] (
17:02:33 Quit stoffel (Ping timeout: 265 seconds)
17:04:29 Quit MethoS- (Ping timeout: 265 seconds)
17:05:42 Join bitcraft [0] (
17:35:38 Join Keripo [0] (
17:46:34***Saving seen data "./dancer.seen"
17:50:49 Quit Keripo (Ping timeout: 256 seconds)
18:18:17 Quit lorenzo92 (Quit: ChatZilla [Firefox 12.0/20120423122928])
18:21:21 Quit TheLemonMan (Quit: Does anybody here remembers of Vera Lynn ?)
18:29:39 Quit Horscht (Quit: Verlassend)
18:41:32 Join liar [0] (
18:44:27 Quit bitcraft (Remote host closed the connection)
19:07:11*pixelma reports more snappy volume control and regular disk access on the Ondio again, thanks for the fix
19:07:44 Quit n1s (Quit: Ex-Chat)
19:08:02 Quit Topy (Read error: Connection reset by peer)
19:08:03 Quit liar (Ping timeout: 252 seconds)
19:09:51 Join liar [0] (
19:24:13 Join lebellium_ [0] (
19:25:16 Quit lebellium (Ping timeout: 246 seconds)
19:25:23 Nick lebellium_ is now known as lebellium (
19:28:23 Nick TotalledZzZz is now known as Totalled (
19:37:36 Join bitcraft [0] (
19:44:53 Join Topy44 [0] (
19:46:38***Saving seen data "./dancer.seen"
19:53:20 Quit bitcraft (Remote host closed the connection)
19:56:36 Quit nosa-j (Read error: Connection reset by peer)
19:56:43 Quit kevku (Ping timeout: 248 seconds)
19:58:02 Join nosa-j [0] (
20:00:39 Quit Totalled (Ping timeout: 244 seconds)
20:03:26 Join Totalled [0] (
20:15:52 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
20:16:04 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
20:31:49bertrikpamaury, getting any result with radio audio yet?
21:06:25 Join stoffel [0] (
21:08:23 Quit y4n (Quit: PÆNTS ØLF!)
21:46:40***Saving seen data "./dancer.seen"
21:49:02 Quit nosa-j (Excess Flood)
21:50:33 Join nosa-j [0] (
22:00:20kugelgevaerts: thanks for fixing
22:00:32 Quit nosa-j (Ping timeout: 245 seconds)
22:19:41 Join max131 [0] (~max131@
22:20:22 Quit bertrik (Remote host closed the connection)
22:22:57 Join bertrik [0] (
22:22:57 Quit bertrik (Changing host)
22:22:58 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
22:24:16 Quit bertrik (Read error: Connection reset by peer)
22:25:34 Join bertrik [0] (
22:25:35 Quit bertrik (Changing host)
22:25:35 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
22:36:07pamaurybertrik: I'm experimenting right now, I was busy this afternoon
22:46:38bertrikI could investigate the zen xfi3 RTC issue or perhaps do some random hacking on other xfi3 sub-systems
22:48:09pamauryperhaps the RTC, I should really give you my ida files
22:48:58CIA-79Commit 3f365fc in rockbox by Thomas Martitz: load_code: Get rid of win32 specific code in favor SDL_LoadFunction & friends APIs.
22:52:03CIA-793f365fc build result: All green
22:58:27CIA-79Commit d523fd6 in rockbox by Thomas Martitz: load_code/overlay: Don't close plugin handle before calling the plugin.
23:00:51bertrikpamaury, pwm is used for backlight fading I suppose, also on the xfi3, right?
23:01:11CIA-79d523fd6 build result: All green
23:01:22pamaurypwm is used for backlight control, fading is done is software I suppose
23:02:29pamauryis there a problem with pwm ?
23:04:13bertrikno, it's just not mentioned on the wiki, but it is shown in the HW info debug menu
23:05:09pamaurythen yeah, it's using PWM. And the button backlight is also using the (somehow undocumented) pwm of the touch chip
23:06:22bertrikwe don't support fading button light yet, I think
23:06:49pamauryno, but I use pwm anyway
23:07:07pamaurysince we support button light brightness
23:07:42bertrikoh we do, I thought button light was either on or off
23:08:06bertrikThe fuze v2 should be able to PWM button light
23:09:54pamaurythe OF does on/off but I managed to get brightness with the pwm
23:10:54bertrikok, I see now our generic backlight code has supported variable button light for a long time
23:11:52pamauryanyway it's hard to see the fading on the button :)
23:13:20bertrikif I had a fuze v2, I'd add configurable button light, but I don't so I won't
23:17:21bertrikif the date/time is wrong, but I enter the date/time config menu, it still shows a reasonable date (may 25 2012)
23:23:42bertrikis rockbox perhaps using the build date as the minimum valid date?
23:24:19pamauryno, I've tried many different dates on the fuze+ with success
23:24:39pamaurywait a minute, I'll see if I can find what the OF does
23:26:12pamaurythe problem is that it's using a persistent register as a minimum value iirc
23:27:04bertrikI see it works now, date/time survives a reboot
23:27:29bertrikbut I've also seen several occasions where date/time got reset
23:28:10pamauryhmmm, only the fuze+ uses this mechanism, the RTC code should be able to set any value on the X-Fi3 (even though it's not consistent with the OF)
23:28:20pamauryperhaps the OF resets it ?
23:28:41pamauryI know the OF on the fuze+ does that if you go before the minimum
23:33:08bertrikcan I still boot the OF by the way?
23:33:59pamauryyes, hold volume down on boot
23:37:38bertrikIt seems that the OF uses 1970 as reference date, not 1980 as currently implemented in the driver
23:38:31bertrikThe OF shows a date in may 28 2002 when rockbox shows may 26 2012
23:40:40 Quit stoffel (Remote host closed the connection)
23:41:01 Quit saratoga (Quit: Page closed)
23:43:04bertrikI doesn't look like the zen xfi 3 uses the offset from the persistent register, because the time difference in equal to 1980/1/1 0:00:00 minus one hour (DST?)
23:43:28bertrikI'll change the date in the OF just to make sure
23:43:28pamauryI'm not so sure, given the OF code
23:44:24pamauryit clearly offset the HW_RTC_SECONDS value with PERSISTENT2
23:44:29 Quit Sundiver_ (Ping timeout: 260 seconds)
23:44:42pamauryI justd didn't find the reference time yet but it's probabyl 1980 or 1970
23:46:20bertrikregister persistent2 appears to contain the offset
23:46:42***Saving seen data "./dancer.seen"
23:46:45bertrikthere's some value in persistent0 and 1 too, but don't know what that means (could be random / undefined)
23:47:16bertrikso it appears to be same as the fuze+
23:47:47pamaurypersistent0 and 1 have special meaning iirc
23:48:00pamaurypersistent2 is clearly used by the OF code in time code
23:48:21pamaurybut it could also be junk code
23:51:09*bertrik compiles and checks
23:51:30pamaurybertrik: can you check if OF time equals persistent2+seconds ?
23:51:33 Join Horscht [0] (
23:51:33 Quit Horscht (Changing host)
23:51:33 Join Horscht [0] (~Horscht@xbmc/user/horscht)
23:52:11bertrikpamaury, yes, it appears so, I set the time forward by 1 day in the OF, then persistent2 contained 1 day - 7 seconds
23:53:12pamauryok, then that's it. It means that rtc_imx233.c should be fixed to use the same trick on the fuze+ and the xfi3
23:53:21pamauryI haven't check on the xfi2
23:55:33 Join lorenzo92 [0] (
23:56:42bertrikfunny how rockbox is so much faster to start and shutdown than the OF
23:58:19pamauryyeah, the OF is just terrible, I don't know what it can be doing during all that time
23:58:41bertrikmeh, it showed an hourglass for about a minute or so

Previous day | Next day