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 2017-02-05

00:00:51 Quit pamaury (Ping timeout: 276 seconds)
00:03:39 Quit lebellium (Quit: ChatZilla 0.9.93 [Firefox 51.0.1/20170125094131])
00:05:13 Join Bilgus_ph [0] (
00:07:57 Join GodEater` [0] (~whoknows@
00:10:17Bilgus_phJhMikeS: so far with pamurys help we have a way to put some data in the firmware from the bootloader right now i'm passing back the volume number i get from my commit on gerrit is that sufficent for your mounting code and or is there any more info you want need or desire?
00:10:39 Quit GodEater (Ping timeout: 240 seconds)
00:14:33jhMikeSit should be. it'll need a path like "dev:/<0>" which can be constructed firmware-side from the volume number
00:16:14jhMikeSthen you'd call root_mount_path with some directory or another to be the root contents
00:20:44jhMikeSnormally the path would be statically declared but if it needs to be dynamic, just provide a function to return it
00:20:45 Quit girafe (Read error: Connection reset by peer)
00:21:13Bilgus_phhow are you handling the original 'boot' drive is it then virtually mounted in the new root? like microsd was?
00:23:14jhMikeSit will be, but hidden from normal directory scans
00:24:37Bilgus_phIs there any reason it needs to stay hidden?
00:24:53jhMikeSno, it's just a flag
00:26:51jhMikeSjust a preference and a way to keep up appearances. paths like /<0>/foo/bar would still be valid whether it's shown or not
00:28:05Bilgus_phoK it'll be tomorow before i'm back to my dev machine i'll finish up the crc code and some error checking then update what I have so far
00:36:44Bilgus_phOh so would that still point back to the internal drive or the boot drive?
00:39:24 Quit ulmutul (Quit: ChatZilla 0.9.93 [Firefox 51.0.1/20170125094131])
00:40:03 Quit Bray90820 ()
00:40:22jhMikeSBilgus_ph: which?
00:40:47jhMikeSthe path would be whatever mounted at point 0 in the root
00:42:47 Quit Bilgus_ph (Ping timeout: 260 seconds)
00:45:46 Quit The_Prospector (Read error: Connection reset by peer)
00:46:11 Join The_Prospector [0] (~The_Prosp@unaffiliated/cornman)
00:46:25 Quit The_Prospector (Client Quit)
00:48:28 Quit alucryd (Remote host closed the connection)
00:49:43 Join alucryd [0] (~quassel@archlinux/developer/alucryd)
00:57:00***Saving seen data "./dancer.seen"
01:06:28 Join Bray90820 [0] (
01:11:47 Quit Senji (Ping timeout: 240 seconds)
01:13:48 Join Senji [0] (~Senji@
01:20:09 Quit skapazzo (Quit: leaving)
01:50:16 Quit Senji (Ping timeout: 245 seconds)
01:50:34 Quit xorly (Ping timeout: 258 seconds)
02:25:48 Quit ZincAlloy (Quit: Leaving.)
02:43:02*__builtin is running into a headache of a bug
02:43:18__builtinbasically the simulator's SDL functions are replacing the ones in the plugin I'm writing
02:43:47__builtinso when the plugin calls SDL_Init(), it ought to call the one in the plugin, but it doesn't and instead calls the host system's SDL
02:43:56__builtindoes anyone know why or how to fix this?
02:57:01***Saving seen data "./dancer.seen"
03:01:14 Quit cc___ (Ping timeout: 256 seconds)
03:09:17 Join The_Prospector [0] (~The_Prosp@unaffiliated/cornman)
03:14:45 Join JimZipCode [0] (448602b4@gateway/web/freenode/ip.
03:15:46 Join TheEaterOfSouls [0] (~chatzilla@unaffiliated/theeaterofsouls)
03:20:47 Quit ender` (Quit: Java is, in many ways, C++--.— Michael Feldman)
03:24:13 Quit JimZipCode (Ping timeout: 260 seconds)
03:45:47 Part TheEaterOfSouls
04:21:49 Quit MrZeus1 (Ping timeout: 255 seconds)
04:37:20 Quit TD-Linux (Ping timeout: 258 seconds)
04:41:51 Join TD-Linux [0] (~Thomas@about/essy/indecisive/TD-Linux)
04:52:30dongsblame the fact that SDL requires dynamic linking
04:53:13dongsa crappy graphics lib needs a dll permanently attached and breaks my code? No thanks!
04:55:40__builtinwhat alternatives are there?
04:57:05***Saving seen data "./dancer.seen"
05:54:50 Quit alexweissman (Remote host closed the connection)
05:55:37 Join alexweissman [0] (
06:08:15 Quit TheSeven (Ping timeout: 256 seconds)
06:08:30 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
06:14:28 Join _mt_ [0] (~MT@2601:482:4402:7b60:181a:dc07:340f:7306)
06:34:37 Quit smoke_fumus (Quit: KVIrc 4.2.0 Equilibrium
06:55:11 Quit furrywolf (Ping timeout: 248 seconds)
06:57:08***Saving seen data "./dancer.seen"
07:55:54 Quit pixelma (Quit: .)
07:55:54 Quit amiconn (Quit: - Chat comfortably. Anywhere.)
07:59:17 Join pixelma [0] (~pixelma@rockbox/staff/pixelma)
07:59:17 Join amiconn [0] (~amiconn@rockbox/developer/amiconn)
08:47:20 Quit _mt_ (Read error: Connection reset by peer)
08:56:15dongs this doesnt do rockbox right
08:56:40dongsah, same as sport.
08:57:12***Saving seen data "./dancer.seen"
09:09:28 Quit Bray90820 (Ping timeout: 276 seconds)
09:18:45 Join paulk-blaze [0] (
09:29:23 Join lebellium [0] (
09:33:08 Quit paulk-blaze (Quit: Leaving)
09:39:33 Join Bray90820 [0] (
09:44:17 Join PurlingNayuki1 [0] (~Thunderbi@
09:45:57 Quit PurlingNayuki (Ping timeout: 264 seconds)
09:46:04 Nick PurlingNayuki1 is now known as PurlingNayuki (~Thunderbi@
09:59:04 Quit PurlingNayuki (Ping timeout: 240 seconds)
10:16:26 Nick GodEater` is now known as GodEater (~whoknows@
10:17:28 Quit GodEater (Changing host)
10:17:29 Join GodEater [0] (~whoknows@rockbox/staff/GodEater)
10:21:24 Join PurlingNayuki [0] (~Thunderbi@
10:21:51 Join edhelas [0] (~edhelas@
10:30:24 Quit PurlingNayuki (Remote host closed the connection)
10:30:33 Join PurlingNayuki [0] (~Thunderbi@
10:31:24 Quit edhelas (Quit: Leaving.)
10:31:37 Join edhelas [0] (~edhelas@
10:39:36 Join ender` [0] (
10:43:37 Quit edhelas (Ping timeout: 245 seconds)
10:57:13***Saving seen data "./dancer.seen"
11:03:27 Join edhelas [0] (~edhelas@
11:06:04 Quit jhMikeS (Ping timeout: 240 seconds)
11:07:59 Quit edhelas (Ping timeout: 248 seconds)
11:12:55 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
11:16:23 Join cc___ [0] (~ac@2001:910:113f:1:6a05:caff:fe1c:1627)
11:18:08 Join edhelas [0] (~edhelas@
11:26:04 Quit edhelas (Ping timeout: 240 seconds)
11:37:18 Join xorly [0] (
11:38:49 Join edhelas [0] (~edhelas@
11:40:03 Quit edhelas (Client Quit)
11:40:12 Join edhelas [0] (~edhelas@
12:08:34 Quit xorly (Ping timeout: 240 seconds)
12:15:57 Quit fs-bluebot (Read error: Connection reset by peer)
12:15:57 Quit bluebrother (Read error: Connection reset by peer)
12:19:06 Quit APLU (Ping timeout: 255 seconds)
12:20:42 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
12:21:55 Quit edhelas (Ping timeout: 258 seconds)
12:22:57 Join APLU [0] (
12:31:43 Join Senji [0] (~Senji@
12:31:57 Join fs-bluebot [0] (
12:31:58 Quit Senji (Read error: Connection reset by peer)
12:32:28 Join Senji [0] (~Senji@
12:57:16***Saving seen data "./dancer.seen"
13:04:09 Quit idonob_ (Ping timeout: 240 seconds)
13:06:43 Join idonob [0] (
13:22:47 Quit idonob (Ping timeout: 240 seconds)
13:25:15 Join idonob [0] (
13:27:39 Quit cc___ (Ping timeout: 255 seconds)
13:43:10 Join skapazzo [0] (~skapazzo@
13:49:53 Join xorly [0] (
13:54:48*pamaury rediscovers the huge simulator mess for buttons
14:00:56 Join MrZeus1 [0] (~MrZeus@2a02:c7f:7008:3400:904b:d1b6:ac78:ce6f)
14:09:30 Quit prof_wolfff (Ping timeout: 255 seconds)
14:24:36 Join prof_wolfff [0] (
14:26:01*dys found a fifth device on the TEAC's SPI bus
14:45:58pamaurydys: what is it ?
14:47:09dysthe display is also wired to it besides the spdif-tx, -rx, adc and the blackfin
14:47:28pamauryon the same bus ?
14:47:49dysja. sck and di are directly connected
14:48:04dysone of the spdif chips has it DO not connected
14:48:35pamauryprobably the lcd has some other dedicated data lines and the spi is just used for control
14:49:42dysit's a small monochrome oled one. maybe it has an internal framebuffer that they write to via spi
14:50:41*pamaury does understand people who put a small monochrome display on such players
14:50:46pamaury*doesn't understand
14:53:42dysi still haven't traced how messages turn up on the display in the dump
14:54:01dysthere seems to be a bigish jump table that obscures things
14:56:13dysthe call graph surrounding the fwupdate code is also quite big. I have traced some of the logic, but not the one that interprets the bytes at the beginning/end of the image
14:57:20***Saving seen data "./dancer.seen"
14:58:27*pamaury tries to understand the kinetic scrolling code and how to integrate touchpad
15:13:47 Join robertd1 [0] (
16:00:51 Join edhelas [0] (~edhelas@
16:00:57dyspamaury: I don't mind the small display. Seeing all those android DAPs on the market, I'd readily trade a HD display for a little more runtime.
16:02:00 Quit MrZeus1 (Ping timeout: 255 seconds)
16:03:50dysotoh, the runtime advertised for the teac (7h) is quite abysmal, considering it has a 3400mAh cell
16:03:56pamaurydys: the screen is not going to make the difference
16:04:10pamaurythere already barely have 5h playback runtime
16:05:00Bilgusdo we have access to the bootloader version from within firmware?
16:05:01 Quit kugel (Ping timeout: 260 seconds)
16:06:05*dys wants the 20h of his iriver H120 back
16:06:38pamauryBilgus: not currently, and the very notion of version is probably not very interesting anyway
16:06:54BilgusI guess the better question is: is there even a versioning system for boot loaders?
16:07:26pamauryBilgus: not really, we usually call them v1, v2, etc, but naming can be arbitrary
16:07:41Bilgusoh ok I was just trying to decide if there was any other relevant data that needed to be pushed back to the firmware
16:07:46pamauryand not stored anywhere except in an (arbitrary) string usually displayed by the bootloader
16:09:51pamauryI can't think of anything right now
16:10:44Bilgusnice thing about your method is that it is trivial to add more
16:11:46pamaurydid jhMikes put somewhere his patch to select root volume ?
16:12:41BilgusNot that I know of I think he is still working on it but he did say he just needed the volume number and that the actual string can be done fw side
16:23:14Bilgusany reason I should leave payload size so large? also I think maybe we should check the payload length to be sure it matches what the boot loader expects
16:25:09pamauryBilgus: maybe you can make it smaller. I would suggest that the bootloader does not check of length equality, but rather fills up to the maximum length
16:25:15pamauryto the forward compatible
16:25:44Bilgusyeah I was just going to say that the payload should never get smaller
16:25:53pamaury(ie if the bootloader wants to put 4 bytes, but the firmware expects 8 bytes, it is still correct, 0 should be a sane default value)
16:26:13pamaurywe will never remove fields, only add some
16:27:01Bilgusyeah and I can styill leave memset since it pulls that back from the struct
16:29:02Bilgusthough I could see where an older firmware with a newer bootloader would cause an issue
16:30:36pamauryBilgus: the design should be forward and backwatd compatible: if the firmware has space more fields, the bootloader fills them with 0. If the firmware does not have enough space, the bootloader fills as many as possible
16:31:03 Join furrywolf [0] (
16:32:56Bilgushmm so ubtract that many bytes from payload size and check if there is enough room I know it doesn't really matter atm but in the future it may
16:41:43 Join StaticAmbience [0] (
16:46:49 Part robertd1
16:55:48pamauryBilgus: I think the simplest approach: 1) have in the bootloader a copy of the structure, *as the bootloader expects it* 2) memset fw's structure with 0 3) copy bootloader structure to fw structure with MIN(bl len, fw len) bytes
16:57:21***Saving seen data "./dancer.seen"
16:58:07Bilgusisn't that going to be a lot more overhead then just keeping track of bytes written?
16:58:48BilgusIdk I guess mem copy isn't terribly inefficient
16:59:52pamaurycopying a few bytes has 0 cost compared to loading the firmware from storage
17:21:24 Join cc___ [0] (~ac@2001:910:113f:1:6a05:caff:fe1c:1627)
17:57:41 Join JanC_ [0] (~janc@lugwv/member/JanC)
17:59:01 Nick JanC is now known as Guest81504 (~janc@lugwv/member/JanC)
17:59:01 Quit Guest81504 (Killed ( (Nickname regained by services)))
17:59:01 Nick JanC_ is now known as JanC (~janc@lugwv/member/JanC)
18:21:47 Quit cc___ (Quit: WeeChat 1.6)
18:27:19 Join jhMikeS [0] (
18:29:09__builtindarn, looking at how the sim loads plugins is weird
18:29:20__builtinit relies on SDL to load... SDL :(
18:34:22pamaury__builtin: I'm not convinced by your sdl port, that's looks like a big library for a very small gain
18:35:58__builtinit has the potential for porting many more programs
18:36:19__builtinnow it just runs a couple graphical demos
18:37:09pamauryyeah but which programs ? the interactions with the user are very limited on a dap
18:37:12furrywolfgrr. technology is getting too small.
18:37:58__builtinthere's a lot of games that only rely on directional keys and an action button
18:38:10__builtine.g. prince of persia
18:38:19furrywolfwhich one of you hid my fuze? :)
18:38:30furrywolffeature request: homing beacon
18:38:48 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
18:39:13__builtinfurrywolf: you can buy an ipod and make the piezo speaker beep periodically ;)
18:40:58 Quit pamaury_ (Remote host closed the connection)
18:44:28 Join kugel [0] (
18:44:29 Quit kugel (Changing host)
18:44:29 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:46:01__builtinI guess I can solve the issue of name collisions by changing all of the names in the ported SDL
18:46:15__builtinwith some objcopy magic, preferably
18:48:14 Quit APLU (Quit: !suicide)
18:49:04Bilguspamaury couldn't we do a union of magic crc and length since magic is useless once the struct is filled?
18:51:43pamauryBilgus: I don't think it is worth it, to save a few bytes
18:52:23pamaurybut as you want
18:52:48Bilgusyeah I suppose it is only 8bytes I suppose if a dap is that close on space it's not going to be booting anything
18:53:04 Join APLU [0] (
18:53:31pamauryor you could decide to merge magic and crc, decide that crc is set to a magic value to find it and then set to a crc, why not
18:53:46 Quit APLU (Client Quit)
18:53:52pamauryBilgus: by the way, don't forgot to fix RoLo as well
18:54:37BilgusI haven't even looked at rolo yet besides a cursory glance
18:55:03pamauryyou basically need to do the same thing as for the bootloader: locate the structure and copy it
18:57:24***Saving seen data "./dancer.seen"
18:59:36furrywolfdoes the fuze have a switch to detect if headphones are connected?
19:00:16furrywolfit seems to have no output with one of my cables, at all, and another I have to pull back out of the jack slightly.
19:02:28pamauryfurrywolf: no, the fuze cannot detect headphones I believe
19:03:04furrywolfhrmm. both channels switch on and off at once, and doesn't make any of the open-ground noises or such bad connections usually make
19:03:23furrywolfbrb, testing more
19:03:24 Join girafe [0] (
19:06:46pamauryfurrywolf: it's to see, go to WPS, plug your headphones, if you get no sound and sound is not paused, your jack is probably dead
19:10:16 Join APLU [0] (
19:21:50__builtin\o/ it works!
19:37:47 Quit pamaury (Remote host closed the connection)
19:53:18Bilguspamaury rolo uses rb-loader > load_firmware so it should be taken care of in that case what are RKW_FORMAT and MI4_FORMAT for?
20:28:46 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
20:31:36 Quit diox (Quit: brb)
20:32:03 Join diox [0] (
20:40:33pamauryBilgus: ok good
20:41:28BilgusI assume those are OF firmware formats? One other issue I can't push my edits to merge back to master is there something special I have to do?
20:42:00pamauryyes rkw is for rockhip and mi4 is for pp
20:42:10pamaurythat's a bit vague ?
20:42:33pamauryBilgus: ^
20:42:41Bilgus ! [remote rejected] HEAD -> refs/for/master (you are not allowed to upload merges)
20:43:11Bilgusmaybe git merge master then push or should I be creating a new commit?
20:47:15pamauryBilgus: you must rebase, we don't use merge
20:47:26pamauryyou must have done a merge at some point
20:47:59BilgusI pulled your patch in I'll figure it out
20:50:06 Quit xorly (Ping timeout: 240 seconds)
20:56:36 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:5dad:6313:b46:9ea5)
20:57:26***Saving seen data "./dancer.seen"
20:57:54 Join xorly [0] (
21:20:33 Quit edhelas (Ping timeout: 240 seconds)
21:20:43BilgusWhat a mess ok Its updated
21:23:02pamauryBilgus: coding style is not consistent: for() but if ()
21:23:22pamauryyou need to adopt whatever coding style rb-loader.c hax
21:24:32 Join mutnai [0] (6db91733@gateway/web/freenode/ip.
21:25:03Bilgusexcess spaces?
21:25:42Bilgusugh ok lol
21:25:50pamauryor lack of, I didn't check what the style of the rest of the file
21:26:10Bilguscan you look over the rest before I do yet another edit?
21:26:24pamauryBilgus: why did you put this declaration in debug menu ?
21:26:24pamaurystruct boot_data_t boot_data; /* needs moved to area of use */
21:26:57pamaurythere is already an extern declaration in bootdata.h
21:27:27 Join TheLemonMan [0] (~root@irssi/staff/TheLemonMan)
21:27:29Bilgushmm It gave me errors when I didn't have it defines?
21:27:47pamaurythat declaration is clearly wrong
21:27:57pamauryit allocates a *new* variable for starters
21:28:10pamauryalso if there was an error you need to understand it
21:28:39pamaurythe code I initially pushed compiled
21:29:20 Quit mutnai (Client Quit)
21:29:52BilgusLet me check it
21:41:22Bilgushell worked fine, weird
21:42:42pamauryBilgus: make clean && make
21:42:46pamauryif you wan to be sure
21:50:39 Quit skapazzo (Ping timeout: 240 seconds)
21:51:29Bilgusok should all be good
21:52:41 Join skapazzo [0] (~skapazzo@
21:53:49 Quit __builtin (Quit: - Chat comfortably. Anywhere.)
21:55:28 Join __builtin [0] (~xray@rockbox/developer/builtin)
21:57:14 Join kylemsguy [0] (
21:59:08 Part __builtin
22:03:57 Join edhelas [0] (~edhelas@
22:14:22 Quit lebellium (Quit: ChatZilla 0.9.93 [Firefox 51.0.1/20170125094131])
22:19:37 Join | [0] (~xray@rockbox/developer/builtin)
22:45:07BilgusI see themes is down again here is a dump of the Theme site for anyone desperate enough to download it (172 MB) (all downloads work from index.htm) if you instead click on the theme and try to download the link needs changed to .zip eg. download(716).htm > download(716).zip
22:46:31|would it be a good idea to have a message on the front page about the themesite/forums?
22:47:08BilgusI guess that all depends on how much longer this will be an issue
22:57:31***Saving seen data "./dancer.seen"
23:18:03 Quit Senji (Ping timeout: 240 seconds)
23:18:27 Quit TorC (Ping timeout: 240 seconds)
23:19:51 Join TorC [0] (~TorC@fsf/member/TorC)
23:20:36 Join Senji [0] (~Senji@
23:26:19 Quit kylemsguy (Quit: Leaving)
23:40:32 Quit TorC (Ping timeout: 256 seconds)
23:41:27 Quit madk (Ping timeout: 255 seconds)
23:41:45 Join madk [0] (~noneofyou@
23:42:28 Join TorC [0] (~TorC@fsf/member/TorC)
23:51:32 Join Senji_ [0] (~Senji@
23:53:47 Quit TheLemonMan (Quit: "It's now safe to turn off your computer.")
23:54:13 Quit Senji (Ping timeout: 258 seconds)

Previous day | Next day