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 2017-02-21

00:29:11 Quit ender` (Quit: The other day Windows told me I might be a victim of software counterfeiting. Funny, I always thought I was a beneficiary.)
00:33:40 Nick dan- is now known as dan_ (~d@freenode/corporate-sponsor/
00:34:10 Nick dan_ is now known as Guest62944 (~d@freenode/corporate-sponsor/
00:38:42 Nick Guest62944 is now known as dan- (~d@freenode/corporate-sponsor/
00:45:15 Quit bastianilso (Quit: bastianilso)
01:05:20***Saving seen data "./dancer.seen"
01:17:26 Join IRCFrEAK [0] (
01:17:26 Part IRCFrEAK
02:11:36 Quit paulk-collins (Quit: Leaving)
02:48:12 Quit MrZeus2 (Ping timeout: 255 seconds)
03:05:12 Join Piece_Maker [0] (
03:05:24***Saving seen data "./dancer.seen"
03:06:43 Quit Acou_Bass (Ping timeout: 258 seconds)
03:06:43 Nick Piece_Maker is now known as Acou_Bass (
04:03:39 Quit cc___ (Ping timeout: 240 seconds)
05:05:25***Saving seen data "./dancer.seen"
06:22:26 Quit Bretos (Ping timeout: 245 seconds)
06:23:26 Join Bretos [0] (
06:39:27 Quit igitoor (Ping timeout: 240 seconds)
06:40:04 Join igitoor [0] (igitur@2a00:d880:3:1::c1ca:a648)
06:45:34 Quit igitoor (Changing host)
06:45:34 Join igitoor [0] (igitur@unaffiliated/contempt)
06:52:38 Quit [7] (Ping timeout: 240 seconds)
06:53:04 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
07:05:28***Saving seen data "./dancer.seen"
07:36:59 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
07:58:42 Quit pamaury (Ping timeout: 255 seconds)
08:14:41 Join parchd [0] (~parchd@unaffiliated/parchd)
08:24:34 Join ender` [0] (
08:46:25 Join petur [0] (~petur@
08:46:25 Quit petur (Changing host)
08:46:25 Join petur [0] (~petur@rockbox/developer/petur)
08:55:20 Join bastianilso [0] (
09:05:31***Saving seen data "./dancer.seen"
09:35:16 Join einhirn [0] (
09:42:28 Quit parchd (Quit: ciao)
10:17:07 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
10:28:47 Quit einhirn (Ping timeout: 260 seconds)
10:30:39 Quit pamaury (Ping timeout: 260 seconds)
10:48:32 Quit petur (Read error: Connection reset by peer)
10:48:50 Join paulk-collins [0] (
10:52:41 Join petur [0] (~petur@rockbox/developer/petur)
10:56:50 Join wodz [0] (
11:01:51wodzThe test points I pointed out yesterday are in did UART (3.3V 115200 8N1).
11:05:35***Saving seen data "./dancer.seen"
11:47:31 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
11:53:04 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:937:e67c:4c2b:cb81)
12:03:03 Quit pamaury (Ping timeout: 260 seconds)
12:05:07 Join dys` [0] (
12:10:15 Quit dys` (Ping timeout: 240 seconds)
12:34:14 Join cc___ [0] (~ac@2001:910:113f:1:6a05:caff:fe1c:1627)
12:45:37 Quit aevin (Ping timeout: 252 seconds)
12:46:03 Join aevin [0] (eivindsy@unaffiliated/aevin)
13:03:33 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
13:05:38***Saving seen data "./dancer.seen"
13:16:41 Join man_in_shack [0] (~chat@unaffiliated/man-in-shack/x-4279753)
13:16:49man_in_shackwelp, my sansa clip has died :(
13:21:59man_in_shackah ok, wasn't holding down power button for long enough for the "reset" to kick in
13:26:34 Quit Rower (Ping timeout: 240 seconds)
13:32:45 Quit wodz (Ping timeout: 240 seconds)
13:59:22 Quit Moarc (Quit: i znowu NADMUCHAƁ BALONA)
14:36:27 Join mutnai [0] (6db91733@gateway/web/freenode/ip.
14:36:52 Quit mutnai (Client Quit)
14:56:15 Quit pamaury (Ping timeout: 240 seconds)
15:05:42***Saving seen data "./dancer.seen"
15:09:45 Join Moarc [0] (
15:18:24 Quit Moarc (Read error: Connection reset by peer)
15:18:50 Join Moarc [0] (
15:30:06man_in_shackhuh. files that play in vlc but not rockbox ...
15:35:01 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
15:37:23 Join Rower [0] (
15:46:50amayerman_in_shack: what type of files?
15:46:58amayerIve had that problem with ogg files
15:47:17man_in_shacklooks like there's an empty theora stream in them
15:47:47man_in_shackthat's probably confusing it
15:48:05amayerman_in_shack: how do you check?
15:50:58man_in_shackyup, that's the problem
15:52:15amayerdid you remove the empty stream?
15:55:48 Quit man_in_shack (Ping timeout: 268 seconds)
15:56:27 Join man_in_shack [0] (~chat@unaffiliated/man-in-shack/x-4279753)
15:58:59man_in_shackamayer: yes
15:59:08man_in_shackwith ffmpeg -vn -acodec copy
16:04:14amayerhow do you just output stream lengths with ogginfo? when I run ogginfo FILENAME it outputs boat loads of data
16:10:07man_in_shackogginfo | grep 'Theora stream' << if that's there, it needs killing
16:17:19amayerman_in_shack: my file didnt contain that.
16:17:28amayerHere is the file I am looking at:
16:19:58ParkerRProtip: If you are simulating your real device be sure to not copy over the codecs... Just spent a few minutes wondering why I couldn't play any media in the simualtor lol
16:23:52ParkerRWtf tried taking a screenshot of the simulator window. It's just black (Linux X11 XFCE maim screenshot tool)
16:25:08 Join parchd [0] (~parchd@unaffiliated/parchd)
16:25:16ParkerRHuh If I do larger than the window it works fine but doing only the window turns out black
16:25:23amayerParkerR: doesnt the simulator have a built in screenshot key?
16:26:51ParkerRDidnt think about that...
16:27:15amayerfirst note in that section
16:27:53amayertoo many # at the end...
16:27:58amayer*drinks more coffee*
16:29:38ParkerRI present: Inverted Rockboxed
16:38:06amayerParkerR: looks cool!
16:52:51 Quit pamaury (Ping timeout: 255 seconds)
17:05:46***Saving seen data "./dancer.seen"
17:07:58 Quit parchd (Quit: working)
17:31:13 Quit alexweissman (Remote host closed the connection)
17:32:09 Join alexweissman [0] (
17:38:33 Quit petur (Quit: Connection reset by beer)
17:38:42 Join wodz [0] (
18:30:07 Quit alexweissman (Remote host closed the connection)
18:31:10 Join alexweissman [0] (
18:36:05 Quit amiconn (Quit: No Ping reply in 64 seconds.)
18:36:46 Join amiconn [0] (~amiconn@rockbox/developer/amiconn)
18:39:28 Join mutnai [0] (6db91733@gateway/web/freenode/ip.
18:56:21wodzpamaury: (log) updated AgptekRocker wiki with my findings
19:05:49***Saving seen data "./dancer.seen"
19:13:15 Quit wodz (Ping timeout: 255 seconds)
19:16:52 Join lebellium [0] (
19:20:01 Join MrZeus2 [0] (~MrZeus@2a02:c7f:7018:6200:88b9:ebcd:d3b5:e1f)
19:36:01 Join parchd [0] (~parchd@unaffiliated/parchd)
19:39:23 Quit amayer (Read error: Connection reset by peer)
19:41:13 Join amayer [0] (
19:53:30 Quit rasher (Quit: leaving)
20:02:58 Quit parchd (Quit: dinner)
20:05:39 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
20:14:28 Quit mutnai (Ping timeout: 260 seconds)
20:15:55 Quit bluebrother (Read error: Connection reset by peer)
20:15:56 Quit fs-bluebot (Read error: Connection reset by peer)
20:21:38 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
20:23:06 Join dys` [0] (
20:32:32 Join fs-bluebot [0] (
20:33:31 Join smoke_fumus [0] (
20:39:49 Join johnb3 [0] (
20:40:08 Join rasher [0] (~rasher@rockbox/developer/rasher)
20:40:32 Quit jhMikeS (Ping timeout: 260 seconds)
20:55:03 Quit johnb3 (Quit: Nettalk6 -
21:05:52***Saving seen data "./dancer.seen"
21:06:08pamaurywodz (logs): thanks
21:23:14BilgusSo If in a plugin I have a static allocation does that then get loaded into the plugin buffer?
21:24:29Bilgusor would I be better off allowing the user to alloc from the buffer lib for say an 8kb buffer?
21:25:21Bilgus@ __builtin or anyone else knowledgeable on the subject of plugins
21:36:56 Join b0hoon [0] (
21:38:42pamauryBilgus: I'm not 100% sure, I think if the buffer is empty, then it's just a bss section and thus you can safely allocate it statically. If the buffer is really big though, better use buflib because you can use part of the audio buf
21:38:48pamauryhopefully someone else can confirm
21:41:33BilgusI think the nice thing about allowing alloc through the buffer lib is less waste if that much isn't needed as well
21:43:22pamaurywell it's not that clear. The plugin bug is reserved for plguins, afaik it is not used for anything else
21:44:13pamauryso assuming it is true, it is in fact better to statically allocate, buflib would borrow from audio buf which would be a waste
21:46:21Bilgusoh ok then size for worst case and just let ride I guess
21:49:54pamauryBilgus: I've looked at the plugin linker file and indeed any static empty buffer will turned in bss and essemtially clear on plugin start using memset. And the plugin buffer is clearly separate from everything else. Just make sure the overall size fits in plugin buf for all platforms
21:50:24Bilgusawesome thanks
21:52:59b0hoonpamaury: have a moment? maybe you could help.
21:58:51pamauryb0hoon: sure
22:02:57b0hoonpamaury: bootloader usb mode is broken in current head, at least for gogear sa9200. Now the bug was introduced or rather exposed by 2a63502
22:03:41b0hoonpamaury: now if you look at it, i get the panic exactly in this place where gevaerts added usb_handle check
22:05:00b0hoonpamaury: so usb_handle is < 0, means no memory, i don't know what's going on, maybe there is no proper buflib init or something for the bootloaders
22:06:33pamauryhum, I know very little about buflib, let me try it on another target just to make sure
22:10:00pamauryb0hoon: does that trigger a panic all the time on sa9200 ?
22:11:21b0hoonpamaury: yes and i wanted to add usb mode for my needs on the samsung yh 820 and the same story
22:11:46pamauryb0hoon: it's a little bit suspicious that apps/ calls core_allocator_init() but none of the bootloader does. However it works like a charm on the fuze+
22:12:20pamauryb0hoon: which file is the bootloader for this target ? (in bootloader/)
22:12:48b0hoonpamaury: oh, so there is init function for bufilib
22:14:18b0hoonthere is no core_allocator_init call there
22:14:54pamauryI see two possibilities: 1) core alloc is somehow not initialised, but then why it does work on other targets? 2) on your target, something else calls core_alloc_maximum before usb
22:17:25pamauryb0hoon: can you try adding core_allocator_init() somewhere in bootloader's main ?
22:17:42pamaurybetween system_init() and kernel_init()
22:18:00b0hoonpamaury: i understand, shit i have no other target than pp to try
22:18:08b0hoonpamaury: sure
22:18:27pamauryI am trying it on the fuze+ to see if it at least causes no regression
22:19:45pamauryah tricky, it doesn't work, because there is no audiobuf
22:20:41pamauryin bootloader, I would expect that it does not use core_alloc at all then
22:21:09pamauryb0hoon: in usb_storage.c:440, there is this conditional
22:21:10pamaury#if (CONFIG_CPU == IMX31L || defined(CPU_TCC77X) || defined(CPU_TCC780X) || \
22:21:10pamaury defined(BOOTLOADER) || CONFIG_CPU == DM320) && !defined(CPU_PP502x)
22:21:56pamaurywhich is a bit weird: it means that in bootloader BUT not on pp502x, it uses use a static buffer, otherwise it uses core_alloc
22:22:01b0hoonoh god
22:22:12pamaurybut core_alloc is ever an option in bootloader so that seems wrong
22:22:23pamauryI am not sure why it is like this
22:22:50pamauryI suspect that if you remove it it will work
22:23:44pamaurybecause if you think about it, CPU_PP502x cannot be set at the same time as CPU_*, so really this is saying BOOTLOADER && !CPU_PP502x in your case which looks really wrong
22:25:48b0hoonyeah the question is why it is like that
22:25:57b0hoonin the first place
22:26:36b0hooni'll try both
22:26:59*pamaury would simply remove the && !defined(CPU_PP502x)
22:29:33 Quit cc___ (Quit: WeeChat 1.6)
22:34:36b0hoonso option with core allocator init doesn't compile: undefined reference to `audiobuffer' and 'audibufend' - like you said there is no audiobuffer
22:36:49b0hoonsecond option doesn't compile either - some errors like: section `.ibss' is not within region `IRAM' and section `.iram' is not within region `IRAM'
22:37:16b0hoonso at this moment i give up
22:38:09 Quit smoke_fumus (Quit: KVIrc 4.2.0 Equilibrium
22:38:26pamauryshouldn't be too hard to fix, maybe the pp must be fixed to include those
22:38:48pamaurygive me 5 min to look into it
22:39:18b0hoonyep however i'm not very familiar with lds files now
22:42:54pamauryhum, it's a bit weird, the pp has two lds files: one for normal bootloader and one for usb bootloader
22:43:05pamaurybut only the normal one handles ibss
22:43:19pamauryI think the usb one needs to be fixed to handle it similarly
22:43:31 Join furrywolf [0] (
22:43:56__builtinBilgus: I believe so
22:44:08__builtin(re. plugin allocations)
22:45:00furrywolfI have an interesting abnormality in the playlist viewer. just created a new playlist. where #39 should be, it instead shows #13833. #13833 is listed again in its proper place.
22:45:41b0hoonMike did the usb bootloader mode, maybe he could look at it
22:47:23Bilgusfurrywolf, are there anymore abnormalities or just that one?
22:48:11furrywolfI didn't spot any more, but I didn't read through 14,000 lines one at a time. :)
22:48:16furrywolfit also breaks playback
22:48:22furrywolfI'm guessing the playlist is corrupted.
22:48:54furrywolfpressing right on #38 causes the player to go out to lunch
22:50:25furrywolfhrmm, after a power off, the playlist is now gone. re-inserting and seeing if it does the same thing.
22:51:05furrywolfyes, it is reproducable.
22:51:55pamauryb0hoon: try g#1569
22:51:56fs-bluebotGerrit review #1569 at : Fix bootloader USB on PP502x by Amaury Pouly
22:52:38Bilgussame place as well?
22:54:47furrywolfif I play it, it plays the correct track sometimes, crashes other times.
22:55:56furrywolf"next track" is blank sometimes, and displays the wrong thing other times.
22:56:21Bilgusok could you run a script to make a number of very small files with names like file_00001.mp3 through whatever the count of files you have is and see if it does the same thing and if not then try it with filenames that are on par with the longest you have?
22:57:04furrywolfI have to go run some errands in a couple minutes... maybe when I get back.
22:57:41Bilguseh whenever..
22:58:05furrywolfif there are other bad entries, they're not obvious, in that they're the same length and the first two digits are good, as that's all I can read when scrolling quickly. :)
23:00:04furrywolfwhen playing, it plays the track that should be in that position
23:00:58 Quit MrZeus2 (Read error: Connection reset by peer)
23:01:03 Quit b0hoon (Ping timeout: 260 seconds)
23:01:45 Join MrZeus2 [0] (~MrZeus@2a02:c7f:7018:6200:88b9:ebcd:d3b5:e1f)
23:01:50 Join b0hoon [0] (
23:02:20 Quit ZincAlloy (Read error: Connection reset by peer)
23:03:52 Quit Rondom (Remote host closed the connection)
23:04:07 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:937:e67c:4c2b:cb81)
23:04:31furrywolfthat is, the actual #39 track, not the #13833 track.
23:04:31furrywolfand this time, with a normal power off, it still restarted with no playlist
23:04:31furrywolfso it's not being saved
23:05:56***Saving seen data "./dancer.seen"
23:06:00 Join Rondom [0] (
23:11:44b0hoonpamaury: "a little better" you could say :) , it compiles and doesn't panic but it hangs on last line "bootloader usb mode", so pc doesn't recognize it of course
23:13:15b0hoonhow the hell it worked in the first place?
23:14:09 Join wodz [0] (
23:19:35pamauryb0hoon: maybe something else is wrong. Tweaking linker script requires knowledge of the platform, but I don't know much about pp
23:20:23pamauryI have to say I am curious to know if there other pp bootloaders with usb work
23:22:35wodzpamaury: I should have b0hoon's Vibe 500 somewhere. But messing with bootloaders sounds dangerous, isn't it?
23:23:33b0hooni suspect that crt0-pp502x-bl-usb.S should be tweaked aswell, but this is just my guess
23:24:19pamauryb0hoon: how does firmware upgrade works on those platforms ? is it easy to recover ?
23:24:28b0hoonwodz: no rather not, you have always original bootloader to restore
23:24:39pamaurywodz: as long as we only touch usb related stuff, we can't get it worse than a panic
23:25:01b0hoonrockbox bootloader is only in the middle
23:25:05pamauryb0hoon: I don't think crt0 needs changes
23:25:13b0hoonpamaury: ok
23:25:36wodzok, then I can look for Vibe 500, charge it and test bootloader if needed
23:26:57b0hoonwodz: but the vibe doesn't have bootloader usb mode, because it's not based on the flash, and you don't need it, i mean it doesn't expose hidden partition to upgrade firmware
23:28:09b0hoonwodz: only gogear sa9200 and some sansa needs it
23:28:24wodzSo I can't help at all.
23:29:02b0hoonwodz: but you can always add it (enable) on any pp target
23:29:26b0hooni think so... :P
23:30:31wodzIf you find it useful I can look for Vibe.
23:31:03wodzIt can't be that hard to find one little player in the drawer full of cables :-)
23:31:10b0hoonwodz: sure
23:33:00b0hoonpamaury: strange thing is it works without the check on usb_handle after all
23:33:45b0hoonand without you patch of course
23:33:50pamauryb0hoon: that sounds like pure luck
23:34:00b0hoonyeah absolutely
23:36:41pamauryb0hoon: I have an idew why it may not be working, wait a minute
23:37:17b0hoonand i'm worng it works with the code from that period i haven't check it with the current head yet without the handle condition
23:37:41b0hoonso maybe another commit broke something
23:39:20pamauryb0hoon: I updated the gerrit task
23:41:17 Join krabador [0] (~krabador@unaffiliated/krabador)
23:42:59lebelliumpamaury: I clicked on "open all" in your NWZ patch. Dozens and dozens tabs opened :D
23:43:27 Nick __builtin is now known as | (~xray@rockbox/developer/builtin)
23:43:35b0hoonpamaury: no luck, it hangs on the tree
23:44:50*pamaury does not know
23:46:09b0hoonpamaury: ok, anyway thanks a lot, i hope jhMikeS reads the logs
23:49:33 Join saratoga [0] (123e1c1b@gateway/web/freenode/ip.
23:49:46saratogajhMikeS: g#1568 looks like an oversight in the voice code
23:49:48fs-bluebotGerrit review #1568 at : voice file not loading by Peter Sealy
23:52:02saratogaalso it would be nice to get AMS bootloaders all building again and then finally update them given the various fixes
23:52:23 Quit saratoga (Client Quit)
23:53:42 Quit alexweissman (Remote host closed the connection)
23:54:07 Quit lebellium (Quit: ChatZilla 0.9.93 [Firefox 51.0.1/20170125094131])
23:55:25 Join alexweissman [0] (
23:57:39 Nick | is now known as __builtin (~xray@rockbox/developer/builtin)
23:58:58 Quit pamaury (Remote host closed the connection)

Previous day | Next day