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 2016-09-05

00:00:46 Quit soap (Ping timeout: 250 seconds)
00:07:02 Quit edhelas (Ping timeout: 252 seconds)
00:15:54 Quit bertrik (Quit: "This incident will be reported")
00:19:23__builtinprof_wolfff: you there?
00:21:27 Quit paulk-collins (Quit: Leaving)
00:28:23prof_wolfff__builtin: yes, here :)
00:29:49__builtinalright, I'm probably in way over my head here
00:29:58__builtinbut I want to take a crack at reverse-engineering the TV out
00:33:06prof_wolfffAFAIK there is not public info about the TV out controller, have the base address annotated somewhere, it has hundreds of registers
00:42:59prof_wolfffin my notes base addresses 0x3910000, 0x3920000, 0x3930000 and clockgates 14,15,16 are marked as "svideo related", i have no more info about it, searching for these ones in OF produces matches on OSOS and diagmode
00:43:28prof_wolfff*0x3910_0000, 0x3920_0000, 0x3930_0000
00:48:31 Join shdwprince [0] (~textual@
00:48:37 Quit ender` (Quit: cannibal, n. someone who's fed up with people)
00:51:42 Quit Roblimo (Read error: No route to host)
00:52:44__builtinooh... "HandleTVSignal"
01:01:01prof_wolfff__builtin: if you are interested i can send you my IDA files, there is nothing about video out, but other low level functions are documented, anyway be aware that almost all my comments are in spanish and some of them are wrong
01:01:34__builtindoes the demo IDA support opening saves?
01:01:47 Join Roblimo [0] (~Roblimo@
01:01:57prof_wolfffopening saves?
01:02:26__builtinsaved project files
01:03:08prof_wolfffi think i am using the demo version and had no problems, but not sure, ATM i have no access to the windows VM to check it
01:04:39 Quit Slasheri (Ping timeout: 244 seconds)
01:06:40__builtinalright, I'll try it
01:07:55prof_wolfffwill try to send it to your email tomorrow
01:08:23*[Saint] wonders if anyone ever paid for IDA
01:08:32 Join Slasheri [0] (miipekk@rockbox/developer/Slasheri)
01:09:03prof_wolfffwell, i tried that, check the price and forget about it
01:20:21Roblimoupdate, if it matters any: I wiped the clip+ with dd if=/dev/zero, and installed the latest firmware
01:20:32Roblimostill no cigar :( - any further suggestions?
01:21:05Roblimo("installed" = dd copy the .bin file)
01:30:15***Saving seen data "./dancer.seen"
01:40:36 Join banjo2 [0] (adbc24ba@gateway/web/freenode/ip.
01:41:28 Quit banjo2 (Client Quit)
01:41:53 Join banjo2 [0] (adbc24ba@gateway/web/freenode/ip.
01:42:02 Quit banjo2 (Client Quit)
01:42:43 Quit xorly (Ping timeout: 255 seconds)
01:43:58 Join Bray90820_ [0] (
01:44:25 Quit Bray90820_ (Read error: Connection reset by peer)
01:45:19 Join Bray90820_ [0] (
01:45:45 Quit Bray90820_ (Read error: Connection reset by peer)
01:46:28 Join Bray90820_ [0] (
01:46:55 Quit Bray90820_ (Read error: Connection reset by peer)
01:47:37RoblimoI guess it's good news that I can at least see a 3.7GB disk with "fdisk -l"
01:47:49 Quit idonob (Ping timeout: 265 seconds)
01:47:52 Join Bray90820_ [0] (
01:48:20 Quit Bray90820_ (Read error: Connection reset by peer)
01:48:20RoblimoIs the only option left to use the JTAG recovery method?
01:49:02 Join Bray90820_ [0] (
01:49:27 Quit Bray90820_ (Read error: Connection reset by peer)
01:50:08 Join Bray90820_ [0] (
01:50:36 Quit Bray90820_ (Read error: Connection reset by peer)
01:53:50[Saint]Roblimo: last time we spoke you were just shitting the firmware image onto the storage it exposed, and not actually the ~900MB "recovery" section, no?
01:54:10[Saint]If that didn't change, then I'm not surprised in the least that your situation hasn't either.
01:54:54[Saint]also - honestly, a JTAG wiggler (one that isn't shit) costs more than a new Clip, even at inflated hipster resale supply/demand prices.
01:55:56Roblimonot sure we have spoken ?
01:56:17[Saint]I think I mentioned this at the time but exposing any section of storage that isn't that ~900 (960-ish from memory) MB section won't help you in the slightest.
01:56:30[Saint]If we haven't, it's an amazing coincidence, but the same rings true.
01:56:58RoblimoDigging around the IRC logs, I found someone with a similar issue back in April
01:57:15[Saint]any section of storage that isn't ~900-odd MB is useless for recovery.
01:57:37RoblimoI get a 3.7GB section of storage
01:57:56Roblimois that useless for recovery?
01:58:20shdwprincecan I ask, in short, how did you turn the mp3 into a brick?
01:59:09[Saint]"shit happens and men are flawed and create flawed designs, made of man" - basically.
01:59:31[Saint]user doesn't need to actually /do/ anything for these devices to implode.
01:59:38shdwprinceI mean, on the technical side
01:59:38RoblimoSomething crashed with rockbox
02:00:10[Saint]that points more towards rockbox crashing as a symptom, more than a cause.
02:00:20Roblimoafter pushing a bunch of the buttons trying to revive it, probably something made it go black
02:00:33[Saint]If we haven
02:00:48shdwprinceI'm kinda used to do all sorts of fw modification stuff - cell phones, players, portable consoles - and never ran into condition where recovery took more than a hour or so
02:00:54[Saint]'t spoken, that's unfortunate, because I wouldn't have advised you mess with dd _at all_.
02:00:56shdwprinceguess I've got lucky
02:01:04[Saint]You probably just needed to disconnect the battery or let it run flat.
02:01:14[Saint]Now...things are a lot more involved.
02:01:37Roblimolol - I zeroed the whole 3.7GB volume with /dev/zero
02:01:44Roblimoand it has a fresh firmware sitting in front :)
02:01:48[Saint]yeah, ouch.
02:02:03shdwprincethere's no official flash app? the one that restore everything and require only flash clip intact to work?
02:02:47[Saint]And the "ahh fuck it, format it" approach is almost always the wrong one with embedded devices.
02:02:51[Saint]But we live and learn.
02:03:09shdwprinceguess it's a good thing that I've went with ipod and not sansa
02:03:38Roblimoso to summarize, I cannot seem to access the 900MB volume
02:03:47[Saint]iPods, for various reasons, are virtually impossible to brick. Even misusing the supplied tools it takes deliberate effort.
02:04:13RoblimoI followed the wiki unbricking tutorial as described, and all I can access is the 3.7GB volume
02:04:14[Saint]very very deliberate effort. they're rather bulletproof.
02:04:36shdwprinceokay, so there's that. and what devices are most fragile of all?
02:04:38[Saint]they're also overengineered and in a significantly higher cost bracket than the Sansas.
02:04:42Roblimoat one point I did get the dredded 4MB volume
02:05:15Roblimobut after following the steps again (which it says to do) I got the 3.7GB volume again
02:05:22[Saint]shdwprince: none you can obtain easily.
02:06:11shdwprincehuh. what about rockbox devices?
02:06:27[Saint]I'm talking about Rockbox devices.
02:06:40shdwprinceoh, sorry, I misunderstood that
02:06:57[Saint]There's several supported devices you don't stand a hope in hell of acquiring.
02:08:21shdwprincedunno why, but that old rockboxes were really stylish-looking
02:08:22[Saint]not for a reasonable price a non-collector would pay, and probably not functional, at least.
02:09:11shdwprinceso, basically, any old device is really easy to brick, and newer ones just got better soft'n
02:09:14shdwprince*'stuff ?
02:09:17[Saint]No one who isn;t a collector is going to pay several hundred dollars for a decade and a half old HDD based player with a few GB storage that likely isn't A: complete, or B: functional
02:09:57[Saint]I wouldn't say "any old device", but a couple of the older ones do happen to be somewhat fragile and irrecoverable without specialist tools, yes.
02:10:16[Saint]but they're self limiting as you'll pretty much never find one.
02:10:36[Saint]even broken or incomplete ones sell for several hundred dollars.
02:12:45[Saint]The Clip model Sansas I basically regard as disposable.
02:13:14[Saint]Though with price gouging due to supply/demand and them being discontinued, that is less and less true these days.
02:14:19shdwprincethey're still good devices tho? havent heard anything bad about them, especially considering the price
02:15:02RoblimoThe sound quality/price for the Clip+ keeps me coming back to them
02:16:48RoblimoSure there are better priced players, and better sounding ones but at the 30ish dollar price point I'm not sure there's anything else
02:21:02 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
02:23:15RoblimoSaint - what would cause an inability to access the 900MB volume?
02:34:45Roblimonevermind - it's dead, fried and now a piece of art
02:35:39shdwprincethat's a shame
02:35:56Roblimofound this:
02:36:11RoblimoSome dude had the same issue, and couldn't get the 900MB volume to show up
02:36:42Roblimoso basically that's a Very Bad Sign (TM)
02:52:23 Join CrashBash-Kun_ [0] (~CrashBash@unaffiliated/crashbash-kun)
02:53:32 Quit CrashBash-Kun (Ping timeout: 260 seconds)
03:04:33 Quit Roblimo (Quit: Leaving)
03:18:19 Quit shdwprince (Quit: My Mac has gone to sleep. ZZZzzz…)
03:30:16***Saving seen data "./dancer.seen"
03:40:31 Quit ZincAlloy (Quit: Leaving.)
03:48:06 Quit krabador (Quit: Leaving)
04:03:40 Quit Moarc (Ping timeout: 265 seconds)
04:04:51 Join Moarc [0] (
04:08:05 Quit prof_wolfff (Ping timeout: 244 seconds)
04:20:58 Join prof_wolfff [0] (
04:31:45 Join idonob [0] (
05:30:20***Saving seen data "./dancer.seen"
05:54:56 Join Bray90820 [0] (
05:55:22 Quit Bray90820 (Read error: Connection reset by peer)
05:56:06 Join Bray90820 [0] (
05:56:33 Quit Bray90820 (Read error: Connection reset by peer)
05:57:21 Join Bray90820 [0] (
05:57:49 Quit Bray90820 (Read error: Connection reset by peer)
05:58:32 Join Bray90820 [0] (
05:58:58 Quit Bray90820 (Read error: Connection reset by peer)
05:59:42 Join Bray90820 [0] (
06:00:10 Quit Bray90820 (Read error: Connection reset by peer)
06:00:58 Join Bray90820 [0] (
06:01:24 Quit Bray90820 (Read error: Connection reset by peer)
06:50:34 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:6d77:7ea6:5da1:9d11)
06:55:08 Quit neersighted (Remote host closed the connection)
06:55:38 Quit CrashBash-Kun_ (Read error: Connection reset by peer)
06:55:44 Join neersighted [0] (
07:26:36 Quit neersighted (Remote host closed the connection)
07:27:10 Join neersighted [0] (
07:29:20 Quit neersighted (Remote host closed the connection)
07:30:21***Saving seen data "./dancer.seen"
07:57:01 Join edhelas [0] (~edhelas@
07:57:33 Quit ZincAlloy (Quit: Leaving.)
08:00:14 Join jeremy1968 [0] (67e08311@gateway/web/freenode/ip.
08:01:31jeremy1968I've been having a bit of trouble, I installed Rockbox for my xduoo x3 on a 2GB microSD card. I've since shifted to a 128GB microSD card, I've copied across the .rockbox folder.
08:01:43jeremy1968But I'm getting a "Bad checksum" error.
08:13:32 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:d0d6:4574:8c99:c8e0)
08:15:03 Quit ZincAlloy (Client Quit)
08:16:25jeremy1968The bad checksum error only occurs with the 128GB card too. The 2GB card still boots fine.
08:25:00 Join einhirn [0] (
08:26:24 Join ender` [0] (
09:03:38 Join petur [0] (~petur@rockbox/developer/petur)
09:07:16 Quit edhelas (Ping timeout: 265 seconds)
09:07:36 Quit Slasheri (Quit: leaving)
09:08:03 Nick Slasherii is now known as Slasheri (
09:08:17 Quit Slasheri (Changing host)
09:08:17 Join Slasheri [0] (~miipekk@rockbox/developer/Slasheri)
09:08:50 Quit jeremy1968 (Ping timeout: 264 seconds)
09:21:26 Quit toli (Ping timeout: 276 seconds)
09:26:28 Join toli [0] (
09:30:25***Saving seen data "./dancer.seen"
09:37:08 Join wodz [0] (
09:46:31 Join elensil [0] (~edhelas@2001:1c02:1903:d800:79b3:93ef:fb43:aa2a)
10:41:01 Join paulk-collins [0] (
10:48:30 Quit pamaury (Ping timeout: 244 seconds)
11:02:08wodzgevaerts: ping
11:15:46 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
11:30:26***Saving seen data "./dancer.seen"
12:08:05 Quit shane (Ping timeout: 244 seconds)
12:08:12 Join shane [0] (
12:11:22 Join shdwprince [0] (~textual@
12:14:52 Join robertd [0] (c9d3b7fd@gateway/web/freenode/ip.
13:06:15 Join krabador [0] (~krabador@unaffiliated/krabador)
13:08:34 Quit robertd (Quit: Page closed)
13:14:54 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:58d4:ffa2:c1a1:eab3)
13:19:15 Join bp0 [0] (~bp@unaffiliated/bp0)
13:30:29***Saving seen data "./dancer.seen"
13:41:59 Join cc___ [0] (~ac@2001:910:113f:1:6a05:caff:fe1c:1627)
14:05:47 Join xorly [0] (~xorly@
14:15:51 Quit xorly (Ping timeout: 240 seconds)
14:28:14 Join xorly [0] (~xorly@
14:37:47 Quit xorly (Ping timeout: 265 seconds)
14:46:11 Quit pamaury (Ping timeout: 244 seconds)
15:24:30 Join [Sinner] [0] (~sinner@rockbox/staff/saint)
15:25:07 Quit [Saint] (Ping timeout: 244 seconds)
15:25:51 Quit wodz (Quit: Leaving)
15:30:26 Quit uwe_ (Ping timeout: 244 seconds)
15:30:33***Saving seen data "./dancer.seen"
15:48:09 Quit __builtin (Ping timeout: 244 seconds)
15:49:41 Join xorly [0] (~xorly@
15:53:21 Join uwe_ [0] (
16:05:51 Quit krabador (Remote host closed the connection)
16:43:57 Quit xorly (Ping timeout: 244 seconds)
16:50:29 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:08:35 Join robertd [0] (c9d3b7fd@gateway/web/freenode/ip.
17:19:28 Quit rudi_s (Read error: Connection reset by peer)
17:20:03 Join rudi_s [0] (
17:21:14 Quit shdwprince (Quit: My Mac has gone to sleep. ZZZzzz…)
17:30:34***Saving seen data "./dancer.seen"
17:39:03 Quit krnlyng (Ping timeout: 260 seconds)
17:43:35 Join ZincAlloy1 [0] (~Adium@2a02:8108:8b80:1700:58d4:ffa2:c1a1:eab3)
17:44:14 Join rela_ [0] (
17:46:25 Join evilnick_ [0] (
17:47:29 Join ruhans_ [0] (uid76353@gateway/web/
17:48:06 Join scorche|1h [0] (~scorche@rockbox/administrator/scorche)
17:48:15 Join user890104_ [0] (Venci@unaffiliated/user890104)
17:48:28 Join PhotoJim_ [0] (
17:48:36 Join GodEater` [0] (~whoknows@
17:48:51 Join uwe_mobile__ [0] (
17:48:56 Quit ZincAlloy (Ping timeout: 260 seconds)
17:48:57 Quit PhotoJim (Ping timeout: 260 seconds)
17:48:58 Quit Staphylo (Ping timeout: 260 seconds)
17:48:58 Quit Rondom (Remote host closed the connection)
17:48:58 Quit rela (Ping timeout: 260 seconds)
17:48:58 Quit ruhans (Ping timeout: 260 seconds)
17:48:58 Quit froggyman (Ping timeout: 260 seconds)
17:48:59 Quit Ivoah (Ping timeout: 260 seconds)
17:48:59 Quit GodEater (Ping timeout: 260 seconds)
17:48:59 Quit preglow (Ping timeout: 260 seconds)
17:48:59 Quit Elfish (Ping timeout: 260 seconds)
17:49:00 Quit ender| (Ping timeout: 260 seconds)
17:49:01 Quit scorche|sh (Ping timeout: 260 seconds)
17:49:01 Quit user890104 (Ping timeout: 260 seconds)
17:49:02 Quit utrack (Ping timeout: 260 seconds)
17:49:02 Quit uwe_mobile (Ping timeout: 260 seconds)
17:49:02 Quit evilnick (Ping timeout: 260 seconds)
17:49:03 Nick user890104_ is now known as user890104 (Venci@unaffiliated/user890104)
17:49:05 Quit petur (Quit: Connection reset by beer)
17:49:13 Join ender| [0] (krneki@2a01:260:4094:1:42:42:42:42)
17:49:35 Join Rondom [0] (
17:49:41 Join utrack [0] (
17:50:16 Quit rela_ (Quit: Leaving)
17:50:42 Join froggyman [0] (~frogs@unaffiliated/froggyman)
17:51:19 Join krnlyng [0] (
17:52:43 Nick ruhans_ is now known as ruhans (uid76353@gateway/web/
17:56:10 Join Elfish [0] (amba@
17:59:12 Join Staphylo [0] (~Staphylo@2a01:4f8:190:126a:d70a:378:c354:a3a3)
18:08:01 Quit Jack87 (Quit: Jack has left the building.)
18:12:27 Join Jack87|Away [0] (Jack87@nasadmin/admin/jack87)
18:12:30 Nick Jack87|Away is now known as Jack87 (Jack87@nasadmin/admin/jack87)
18:13:36 Quit Jack87 (Client Quit)
18:16:31 Join Jack87 [0] (Jack87@nasadmin/admin/jack87)
18:47:44 Join preglow [0] (~thomj@2001:840:4243:3::101)
19:07:57 Join edhelas [0] (~edhelas@
19:19:43 Quit [7] (Ping timeout: 258 seconds)
19:20:02 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
19:27:03 Join shdwprince [0] (~textual@
19:27:17 Join Ivoah [0] (uid49352@gateway/web/
19:30:38***Saving seen data "./dancer.seen"
19:34:41 Join rela [0] (~x@pdpc/supporter/active/rela)
19:43:09 Join lebellium [0] (
20:01:06 Quit quaz0r (Ping timeout: 252 seconds)
20:01:41 Quit edhelas (Ping timeout: 276 seconds)
20:02:44 Quit robertd (Quit: Page closed)
20:14:04 Join quaz0r [0] (
20:17:45 Join petur [0] (~petur@rockbox/developer/petur)
20:41:28 Join wodz [0] (
20:41:49wodzgevaerts: ping again
20:53:48 Join CaptainKewl [0] (
20:59:39 Quit CaptainKewl (Quit: ( :: NoNameScript 4.22 :: ))
21:00:45 Join CaptainKewl [0] (
21:05:06 Part CaptainKewl
21:05:08 Join CaptainKewl [0] (
21:14:50 Join Hoshi [0] (
21:15:01Hoshipamaury: how the nwz port is going
21:30:30 Join hoshi_ [0] (
21:30:41***Saving seen data "./dancer.seen"
21:46:32 Quit Hoshi (Quit: Leaving)
21:47:23 Join Bray90820 [0] (
21:51:00 Join girafe2 [0] (
21:53:14 Quit girafe (Ping timeout: 264 seconds)
21:56:16 Join edhelas [0] (~edhelas@
21:59:00 Join robertd [0] (c9d3b7fd@gateway/web/freenode/ip.
22:10:12 Quit Ivoah (*.net *.split)
22:10:12 Quit Elfish (*.net *.split)
22:10:18 Quit edhelas (Ping timeout: 260 seconds)
22:10:51 Join Elfish [0] (amba@2001:1608:12:1:13:3:3:7)
22:11:51 Join Ivoah [0] (uid49352@gateway/web/
22:14:03gevaertswodz: pong?
22:14:30 Quit Ivoah (Ping timeout: 251 seconds)
22:14:52 Join __builtin [0] (~xray@unaffiliated/franklin)
22:16:41wodzgevaerts: Two things. 1) I am porting picoTTS as plugin. Uploaded POC as g#1389. Now what to do with lingware files? Debian provides it as non-free since the tools to build binary resources are binary only windows exec BUT actual binary resources are considered to freely distributable.
22:16:43fs-bluebot_Gerrit review #1389 at : Port of picoTTS by Marcin Bukat
22:17:41wodz2) If we consider this binary files distributable where to hack makefiles to install some binary resources somewhere (for plugin)?
22:20:06 Quit Bray90820 ()
22:20:24gevaertsI'd say maybe the best way is to have rbutil grab them and install them
22:21:29 Join Ivoah [0] (uid49352@gateway/web/
22:23:23wodzgevaerts: Our standard approach was to instruct users to download resources on their own (doom, rockboy, etc.). This is a bit different as lingware data are considered free from the point of distribution and non free from the point of view of Free Software rules
22:24:18gevaertswodz: I'm not sure how this differs from the doom shareware data, really
22:25:58wodzgevaerts: subtly :-) The source of lingware is provided under permissive license 'only' build tools are not
22:26:35gevaertsHow big are these files?
22:26:44wodzgevaerts: binary or source?
22:26:53__builtinwodz: awesome!
22:27:15gevaertsAs in, how much of an extra install are they for those who don't want them?
22:27:33wodzgevaerts: 6.2MB
22:28:07__builtincan I help test it?
22:28:25wodz__builtin: sure feel free
22:28:41gevaertsHmmm, I'd say that's big enough to have a separate download
22:28:53 Join edhelas [0] (~edhelas@
22:28:57gevaertsI mean, that's twice the size of the fontpack that we do keep separate :)
22:29:08__builtinare the resources compressed?
22:30:07wodzgevaerts: Its about 1MB per language and provided resources are for en-GB, en-US, de-DE, fr-FR, es-ES and it-IT
22:30:07 Nick [Sinner] is now known as [Saint] (~sinner@rockbox/staff/saint)
22:30:14gevaertsThey seem to compress to about 4.5MB here
22:30:58 Quit shdwprince (Quit: My Mac has gone to sleep. ZZZzzz…)
22:32:19[Saint]it's only 1/10th the size of /my/ fontpack. :p
22:32:46*gevaerts thinks
22:32:54[Saint]hmmm, actually, I'm not sure I ever updated the link on the extras page.
22:33:03[Saint]I think that's considerably older/smaller.
22:33:36gevaertsActually, there's also the GPLv2 vs Apache2 we need to take into account
22:33:52gevaertsI'm still of the opinion that as we're GPLv2*+*, it
22:34:03gevaertss not an actual practical problem as such
22:34:11wodzApache2 is compatible with GPLv2, no?
22:34:24gevaertsApache says it is, FSF says it isn't
22:34:32[Saint]FSF says a lot of things.
22:35:11[Saint]FSF's basic beef is that they take exception to the fact that Apache /can/ be closed, right?
22:35:17gevaertsBut as we're GPL2 or later, and GPL3 definitely *is* compatible, we can release binaries as GPLv3 while keeping (most) source GPLv2 for those who need GPLv2 compatibility for some reason
22:36:57gevaertsI'd say it would be nice to have a configure option to disable v3-only bits though. For a plugin, that really just means one extra #ifdef in SOURCES and a tiny bit in configure
22:37:03[Saint]Letting in Apache sources does require some considerations. But I'm not sure how practical they are.
22:37:28[Saint]Someone could be an almighty douche and fork and only release the GPL aspects and break the world.
22:38:01wodz[Saint]: Maybe I read it wrong but Apache license is basically BSD
22:38:43[Saint]doesn't GPL prevent other restrictions on redistribution, though?
22:39:01[Saint]and someone being able to go "nah, this bit just isn't open today" sounds like a restriction.
22:39:07gevaertsGPL prevents extra restrictions
22:39:27gevaertsAnd apache2 has some patent ternmination clauses, while GPLv2 doesn't say anything about that
22:39:40gevaertsGPLv3 does though, apparently in a non-conflicting way
22:40:20[Saint]Just offhand, I don't /think/ we can do it.
22:40:25[Saint]But IANAL.
22:40:30gevaerts[Saint]: why not?
22:40:55wodzI apparently stopped understanding FLOSS licenses after GPLv2
22:41:26*gevaerts thinks it's fine and the resulting mess is only slightly messy :)
22:41:56[Saint]gevaerts: I guess it's the definition of what restrictions are, perhaps it's a personal thing or I'm misparsing it, but incorporating Apache gives others, even us, the ability to restrict those aspects - and I'm not sure how that sits with GPLv2.
22:42:10__builtinwodz: how do I get it to run?
22:42:15[Saint]Someone, including us, could for whatever reason just say "not today" for the Apache bits and lock up house.
22:42:23[Saint]Which even if it is "right" feels very wrong.
22:42:35gevaerts[Saint]: how?
22:42:46gevaerts"Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work and such Derivative Works in Source or Object form"
22:42:47wodz__builtin: it is viewer. Open context menu on .txt file and select open with...
22:42:59gevaerts"irrevocable" is fairly unambiguous I'd say
22:43:43__builtinit seems to lock up for me
22:44:02[Saint]that's only IFF you release the source, (no?) which you're not required to do, by my understanding.
22:44:14[Saint]see: Android - the only ones I know of who ever tested this
22:44:17gevaertsAre we planning not to release the source?
22:44:34[Saint]Someone else may not. Which is where (I think) this gets messy.
22:44:50[Saint]I'm thinking down the chain...not necessarily us.
22:45:39gevaertsIf someone else does that, they either have to strip out all the GPL source first (in which case it's picotts they're distributing, not rockbox), or they have to violate the gpl
22:45:42[Saint]To me that sounds very much like an added restriction. Even if it is optional.
22:47:25[Saint]But, does it violate the GPL? Android kernel springs to mind. Android ended up a multilicense GPL/Apache distributed work and they sure managed to withhold the Apache aspects.
22:47:39[Saint]Albeit precisely once.
22:47:57*gevaerts is a bit confused
22:48:21gevaertsAll involved parties agree that the kernel/userspace split there is *not* "linking" in the GPL sense
22:48:31gevaertsSo I don't see how that's relevant
22:49:34wodz__builtin: have you copied lingware files to the right place? (.rockbox/lang_pico dir)
22:49:44[Saint]Hmmm, that's a fair point. I guess I forgot that I didn't agree with their definition of linking/distributed works.
22:50:09gevaertsAlso, as far as I can see your argument would apply equally to the BSD licenses, and we *do* have plenty of source under those
22:50:18 Join shdwprince [0] (~textual@
22:50:47[Saint]Yes. i do seem to recall having that discussion with you at some point many moons ago.
22:51:06[Saint]I think when we went GPLv2.
22:51:15gevaertsplus, presumably :)
22:51:21[Saint]Ah, yes.
22:52:12[Saint]So, yes. OK. I concede. We can get away with it. It feels dirty though. But I now understand that's for largely personal reasons and not legal reasons.
22:52:34__builtinwodz: it also crashes on hardware
22:52:46__builtin"undefined instruction"
22:52:53wodz__builtin: which hardware?
22:53:19 Quit shdwprince (Client Quit)
22:53:20*__builtin will try to debug
22:53:22wodz__builtin: interesting
22:56:42wodzSo the BIG question is: Can we include Apache2 licensed code or not? Plugins are linked to core technically and hence are linked with GPL2.
22:57:42gevaertsI say "yes", but if we do the resulting *binaries* have to be GPLv3+, not GPLv2+
22:57:56gevaertsWhich really doesn't change anything at all
22:59:35gevaertsPersonally I'd add a configure switch to enable building in fully GPLv2-compatible mode, but that would not be the default, and it really is just about nitpicking
22:59:40__builtinwodz: ok, I've figured out that it crashes in pico_initialize
23:00:25[Saint]A GPLv2 compat build flag would be trivial.
23:00:50gevaertsWell, as long as the non-compatible bits can be left out fairly easily anyway
23:01:22gevaertsFor a plugin it's trivial, but if it's the ATA driver on some target it would be a bit annoying
23:01:23[Saint]I would opt to add that as an axcercise for the user normally, but it would be so trivial to include that...well, why not. Basically.
23:01:49gevaertsBut for the case we're discussing, it's really a non-issue
23:02:21[Saint]I'd wager a majority of people don't even use voice and wouldn't notice wither way.
23:03:01__builtinwodz: the error seems to be non-deterministic
23:03:28__builtina.k.a. it changes each time
23:03:31gevaertsI imagine this might evolve into a core (non-plugin) thing, but even then building without TTS almost certainly is something we want to allow for binsize reasons even if it's not about licenses
23:03:59 Join Bray90820 [0] (
23:04:02 Quit lebellium (Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617])
23:04:07[Saint]I kinda like the idea of the voicing system being modular.
23:04:21wodz__builtin: Could be memory alignment. Try to load txt file which has filesize % 4 = 0
23:04:23[Saint]I /think/ that's an aspect wodz seems to enjoy as well.
23:04:40[Saint]The idea that the voicing system could be easily plugged in and out is appealing.
23:04:57gevaertsWell, I still want relocatable plugins for that sort of thing, but someone has to actually *do* it :)
23:05:43[Saint]wodz: presumably that was something you were shooting for, or is that a happy accident?
23:05:58[Saint]re: modular pluggable voicing system
23:06:00gevaertsIf it's just the one module you want it for, you can link it at the end (or start) of the audio buffer and handle it that way
23:06:16gevaertsFairly trivial, but also fairly non-flexible if people want more stuff to be modular
23:06:41__builtinwodz: still crashes, this time with a "data abort"
23:07:00__builtinI've got it narrowed down a bit more
23:08:03wodz[Saint]: I started to work on relocatable plugins because I tried to port flite TTS back then. The final was I did POC for relocatable plugins and lost interest in porting flite :-)
23:08:56*[Saint] nods
23:10:05__builtinwodz: alright, it's on the line 'sysMM = picoos_newMemoryManager(rest_mem, rest_mem_size, enableMemProt ? TRUE : FALSE);' in libsvox/picoapi.c
23:10:16__builtinaround line 105
23:15:54wodz__builtin: I still think this due to alignment. I didn't pay much attention to this. Will look later.
23:30:42***Saving seen data "./dancer.seen"
23:32:50 Quit wodz (Ping timeout: 265 seconds)
23:49:57 Quit alexweissman (Remote host closed the connection)
23:50:38 Join alexweissman [0] (
23:50:41 Quit edhelas (Ping timeout: 255 seconds)

Previous day | Next day