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 2016-09-21

00:21:43 Quit edhelas (Ping timeout: 250 seconds)
00:29:20 Quit The_Prospector (Read error: Connection reset by peer)
00:32:05 Quit xorly (Ping timeout: 255 seconds)
00:35:08 Join The_Prospector [0] (~The_Prosp@unaffiliated/cornman)
00:42:08 Quit ZincAlloy (Quit: Leaving.)
00:59:21 Quit girafe2 (Read error: Connection reset by peer)
01:16:58 Join alexweissman [0] (
01:20:22***Saving seen data "./dancer.seen"
01:29:30fs-bluebot_Build Server message: New build round started. Revision 9dc4b00, 255 builds, 15 clients.
01:39:11fs-bluebot_Build Server message: Build round completed after 581 seconds.
01:39:12fs-bluebot_Build Server message: Revision 9dc4b00 result: All green
01:39:13fs-bluebot_Build Server message: New build round started. Revision 5e2600e, 255 builds, 15 clients.
01:39:31[Saint]We really need to figure out which method we want to commit in order to work around modern gcc in
01:39:51[Saint]My method, pamaury's method, [7]'s method, etc.
01:40:25[Saint]I opt for mine, personally - but...I would.
01:40:43pamaury[Saint]: [7] method is better but I am holding back because I am lazy and I want to commit the mips toolchain fix at the same time. But that requires switch a toolchain which is a lot of work
01:40:51pamaurywhat is your method?
01:44:06pamaury[Saint]: ^
01:44:30pamauryMy method is a gcc patch. [7] is a MAKE flag and yours is?
01:48:08[Saint]Mine is changing CFLAGS to use -fgnu89-inline
01:48:30fs-bluebot_Build Server message: Build round completed after 558 seconds.
01:48:31fs-bluebot_Build Server message: Revision 5e2600e result: All green
01:48:32[Saint]I do this directly in, this is likely the simplest approach, IMO.
01:49:18[Saint]just s/CFLAGS=-U_FORTIFY_SOURCE ../$toolname-$version/configure/CFLAGS="-U_FORTIFY_SOURCE -fgnu89-inline" ../$toolname-$version/configure/g
01:49:40[Saint]somewhere around line 21* in
01:49:57[Saint]No makefile fuckery, no gcc patching.
01:50:22[Saint]pamaury: ^
01:50:41[Saint](sorry, I had to run off and deal with a cat puking on my rug :P)
01:50:50pamaurythat's [7] method
01:50:57pamaurybut it breaks the mips toolchain
01:51:10pamaurybecause it's so old it doesn't know about -fgnu89-inline
01:51:57[Saint]IIRC, [7] was doing this at the makefile level - his approach was functionally similar, but mine was a little
01:53:10pamauryah ok, anyway what you described is what I had in mind
01:53:22pamauryexcept for this problem with mips toolchain
01:53:54[Saint]I haven't played too much with the new mipsel toolchain, though I do recall it being a problem, in so far as fixing mipsel breaks armeabi and vice versa.
01:54:55[Saint]I ended up in a state where, basically, with a bleeding edge Arch or Debian system I could compile one or the other - but not both.
01:55:58[Saint]I think I should finish and commit for review my installation wrapper for prebuilt toolchains.
01:56:44[Saint]Realistically, I don't think end users, even if they intend on modifying Rockbox and compiling their own builds, have any business building and rebuilding toolchains that haven't (and won't) been changed in a decade.
01:57:35[Saint]The thing holding my back is that I couldn't really test for or test on esoteric architectures and operating systems.
01:58:41[Saint]The little wrapper I built was very simple and could only deal with x86/x86_64 and it assumed the distribution was a Debian derivative.
01:59:46[Saint]I understand that being able to compile the toolchains is a very important thing, but I don't see any reason why end users or those wanting to join the build farm should have to do so.
02:00:20 Quit alexweissman (Remote host closed the connection)
02:04:09pamauryI don't have any objection to giving prebuilt toolchain. I just want to make sure we can always build them from source if we need (which is not the case currently anyway)
02:04:16pamaurybedtime for me
02:04:37[Saint]I would like to get these problematic toolchains armeabi/mipsel/arm-ypr0 to compile against modern build tools, and then compile them for x86/x86_64/armel/armhf/aarch64/etc. and just provide a wrapper to install them.
02:05:15[Saint]I'm perfectly willing to host them. I doubt it will be high traffic.
02:05:51[Saint]Though I understand there's some other project I forget the name of that is using our armeabi toolchain.
02:06:27[Saint]For me, honestly, the biggest problem in the toolchains is the fucking arm-ypr0 toolchain.
02:06:33pamauryI was thinking about YP-R0. There is a simple fix for it which would be 1) don't build an dynamically linked executable (ie build a static one) so we don't depend on the system 2) change the toolchain version so that it compiles.
02:06:33pamauryWe cannot do 2) without 1) because currently we need to match at least glibc, but I don't see why we really want to do that (except maybe to same RAM).
02:06:34 Join alexweissman [0] (
02:07:28pamauryI least on the Sony NWZ linux port that I am planning to do, that's what I am doing. This solves all the problem and we don't have to build the crazy sony toolchain
02:07:46pamauryok bedtime for real now, but you can think about it
02:08:56 Quit alexweissman (Remote host closed the connection)
02:17:38 Quit pamaury (Ping timeout: 260 seconds)
02:20:54 Quit __builtin (Quit: - Chat comfortably. Anywhere.)
02:22:43 Join __builtin [0] (~xray@unaffiliated/franklin)
02:29:13 Quit __builtin (Quit: - Chat comfortably. Anywhere.)
02:30:14 Join __builtin [0] (~xray@unaffiliated/franklin)
02:46:46 Quit duo8 (Ping timeout: 276 seconds)
02:47:03 Join duo8 [0] (~ZNC-SRV-H@
02:52:43 Quit robertd (Quit: Page closed)
03:09:56 Quit duo8 (Ping timeout: 244 seconds)
03:17:48 Join duo8 [0] (~ZNC-SRV-H@
03:19:15 Join alexweissman [0] (
03:20:23***Saving seen data "./dancer.seen"
03:37:13 Join saratoga [0] (497221ff@gateway/web/freenode/ip.
03:38:46saratogapamaury: can you promote the imx targets to stable? i think the only thing missing is to tag a bootloader in git
03:48:22 Quit saratoga (Quit: Page closed)
03:49:42 Join shamus [0] (~shmaus@
04:03:40 Quit duo8 (Ping timeout: 265 seconds)
04:07:49 Join duo8 [0] (~ZNC-SRV-H@
04:15:02 Quit alexweissman (Remote host closed the connection)
04:19:57 Join alexweissman [0] (
04:27:13 Nick Strife89|Quassel is now known as Strife89 (
04:48:32 Quit akaWolf (Ping timeout: 240 seconds)
05:00:08 Quit alexweissman (Remote host closed the connection)
05:01:32 Join alexweissman [0] (
05:13:51 Quit smoke_fumus (Quit: KVIrc 4.2.0 Equilibrium
05:20:24***Saving seen data "./dancer.seen"
06:10:59 Quit __builtin (Ping timeout: 260 seconds)
06:42:38 Quit TheSeven (Ping timeout: 250 seconds)
06:43:08 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:50:30 Quit idonob (Ping timeout: 265 seconds)
06:55:45 Join idonob [0] (
07:10:20 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:e58a:263c:68c3:cd4d)
07:20:27***Saving seen data "./dancer.seen"
07:28:54 Quit Bray90820 ()
07:55:20 Quit ZincAlloy (Quit: Leaving.)
07:55:24 Join einhirn [0] (
07:57:53 Quit bluebrother (Disconnected by services)
07:57:58 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
07:59:04 Join fs-bluebot [0] (
08:01:10 Quit fs-bluebot_ (Ping timeout: 244 seconds)
08:02:16 Join ender` [0] (
08:13:29 Quit pixelma (Quit: .)
08:13:29 Quit amiconn (Quit: - Chat comfortably. Anywhere.)
08:16:04 Join pixelma [0] (~pixelma@rockbox/staff/pixelma)
08:16:05 Join amiconn [0] (~amiconn@rockbox/developer/amiconn)
08:19:15 Join edhelas [0] (~edhelas@
08:28:22 Quit edhelas (Ping timeout: 250 seconds)
08:43:23 Join xorly [0] (
08:44:53 Join edhelas [0] (~edhelas@
08:48:16 Join wodz [0] (
09:05:44 Quit edhelas (Ping timeout: 248 seconds)
09:17:20 Join petur [0] (~petur@rockbox/developer/petur)
09:20:28***Saving seen data "./dancer.seen"
09:21:28 Quit tchan (Ping timeout: 240 seconds)
09:24:08 Quit petur (Ping timeout: 265 seconds)
09:24:34 Join petur [0] (
09:24:34 Quit petur (Changing host)
09:24:34 Join petur [0] (~petur@rockbox/developer/petur)
09:27:46 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
09:37:22 Quit xorly (Ping timeout: 272 seconds)
09:43:32 Join elensil [0] (~edhelas@2001:1c02:1903:d800:1cee:8aad:dc3e:84ce)
10:12:52 Join MrZeus_ [0] (~MrZeus@
10:27:28 Quit hoshi (Ping timeout: 240 seconds)
10:30:15 Join goom [0] (
10:30:25 Join hoshi [0] (
10:42:29 Quit goom (Quit: Leaving)
11:06:04 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
11:08:24 Quit JdGordon (Ping timeout: 248 seconds)
11:20:31***Saving seen data "./dancer.seen"
11:20:41 Quit olspookishmagus (Ping timeout: 265 seconds)
11:20:48 Join olspookishmagus [0] (
12:13:42 Join paulk-collins [0] (
12:15:36 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
12:19:30 Join robertd [0] (c9d3b7fd@gateway/web/freenode/ip.
12:25:17pamaurysaratoga (logs): yes I will do that tonight, and also try to have a manual for those too
12:33:38 Quit pamaury (Ping timeout: 260 seconds)
12:34:14 Join JdGordon [0] (~jonno@
12:34:14 Quit JdGordon (Changing host)
12:34:14 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
12:37:03 Quit JdGordon_ (Ping timeout: 244 seconds)
12:43:06 Join smoke_fumus [0] (~smoke_fum@
12:50:37 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
13:01:31 Quit shamus (Quit: chaos reigns within reflect repent and reboot order shall return)
13:03:26 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
13:06:02 Quit JdGordon (Ping timeout: 240 seconds)
13:20:35***Saving seen data "./dancer.seen"
13:33:20 Quit paulk-collins (Remote host closed the connection)
13:48:59 Join xorly [0] (~xorly@
13:56:28 Quit xorly (Ping timeout: 260 seconds)
13:58:04 Join xorly [0] (~xorly@
14:10:38 Quit Strife89 (Ping timeout: 255 seconds)
14:10:50 Join Strife89 [0] (
14:16:18 Quit petur (Read error: Connection reset by peer)
14:16:44 Join petur [0] (
14:16:45 Quit petur (Changing host)
14:16:45 Join petur [0] (~petur@rockbox/developer/petur)
14:23:18 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:d536:6cd5:f64:2e9f)
14:35:25 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
14:40:17 Join p3tur [0] (~petur@rockbox/developer/petur)
14:42:42 Quit petur (Ping timeout: 265 seconds)
14:42:52 Nick p3tur is now known as petur (~petur@rockbox/developer/petur)
15:16:11 Quit duo8 (Ping timeout: 244 seconds)
15:19:02 Join duo8 [0] (~ZNC-SRV-H@
15:19:43 Nick MrZeus_ is now known as MrZeus (~MrZeus@
15:20:37***Saving seen data "./dancer.seen"
15:25:42 Join paulk-veyron-min [0] (~paulk@
15:35:38 Quit wodz (Ping timeout: 260 seconds)
15:44:17 Quit paulk-veyron-min (Quit: Leaving)
16:46:22 Quit petur (Read error: Connection reset by peer)
17:05:28 Quit nlogex (Ping timeout: 260 seconds)
17:08:25 Quit idonob (Read error: Connection reset by peer)
17:08:40 Join idonob [0] (
17:17:00 Join saratoga [0] (123e11e0@gateway/web/freenode/ip.
17:17:15saratogapamaury: sounds great, keep in mind we've never required a manual before to promote to unstable/stable
17:17:28saratogaso you don't need to wait for it to be done
17:17:59pamauryI know but most of them are easy targets and I'd like them to be done. At least the NZZ and also the ZEN X-Fi3
17:19:18pamauryAlso I have a question: I recnrly did some work on the STFM1000 tuner (found in NWZ and ZEN X-Fi3). It's working (at least basic playback, not recording or RDS). Do you think I should clean it up for the release or that can wait?
17:19:48saratogawe haven't really decided on a release date
17:19:57saratogafor 3.14
17:20:16pamauryok, I was thinking to clean up and commit. It cannot release hurt the release (since current state is not working and that would be maybe-work ;))
17:20:38saratogaif you're worried about one specific device, at least promote to unstable
17:20:41***Saving seen data "./dancer.seen"
17:20:51saratogaright now we're probably discouraging a lot of users who would be fine with the current port
17:21:07pamauryyeah, I don't think the radio should be an item against stable
17:21:19saratogathe idea of stable/unstable was to let users know if a port could be used to play music reasonably well
17:21:29saratogawe probably have like 10 devices that fit that label but aren't listed
17:21:33pamauryI mean this radio tuner is incredibly hard to make it work, we'll never make it to stable if want full support
17:21:48pamauryalso I updated the wiki with some bug/enhancements items
17:21:59saratogawe never intended stable to mean that everything worked perfectly, just that you could install rockbox on the device and reasonably play music
17:22:29saratogaunstable meant that and also you would have to be reasonably familiar with the command line or other tricks to install
17:22:30pamaurybasically I did git log −−author=me and translated that into human form
17:22:45saratogayeah i saw, thanks
17:23:17gevaertsWe do have some ports where we support the FM tuner even if the OF doesn't (in some regions, anyway). This would merely balance things out :)
17:23:54pamauryhum by the way, who knows DSP and FM/RDS in particular? saratoga maybe?
17:24:05saratoganot very much
17:24:14gevaertsI know bertrik has worked on rds
17:24:22pamauryI know but it's low-level RDS
17:24:32saratogai looked at an RDS driver that was opens ource a while back just because i was curious about how they do the demoluation
17:24:33pamauryI have to do the demodulation in software
17:24:48saratogacan you use the linux one (or wherever it was from)?
17:24:53pamauryI have the OF code but it's incredibly complex
17:24:56saratogawriting those filters is probably painful
17:25:24pamauryI would prefer to understand how it works ^^
17:27:43pamauryI never realised before that even stereo radio is a pain, you cannot simply use unfiltered audio, otherwise it sounds horrible
17:28:34saratogaha yeah
17:28:37saratogaFigure 2
17:30:33pamauryyeah I looked at that, I'm just not familar enough with DSP so that it really makes sense except at the theoretical level
17:31:54saratogaits a little different than the kinds of things i've done as well
17:32:21saratogabut the basic idea is that you have some carrier above the music, and from that you can recover a clock, then you look to see if there is a high-low transistion on clock edges or not
17:32:41saratogathat gives you the data stream, and then you have a whole mess of the data stream format above that
17:32:50 Join nlogex [0] (
17:33:05pamauryAlso I am not entirely sure which part of the diagram is implement by the tuner. It basically feeds me with the RDS signal at 38kS/s which is (i guess) the output of an ADC at (1) in the diagram of figure 2
17:34:42saratogai guess it is probably 1, i don't think anything else would make sense
17:35:27saratogai wonder if there is some local FM transmittor that supports RDS to compare to
17:35:31saratogalike for an ipod or whatever
17:36:33saratogathis would have been a great class project in college
17:37:02pamaurybut for example, when I looked at the sample (that was a while back), it was a little weird: the amplitude was going up and down (to 0) very slowly, suggesting a small drift between the tuner recovered subcarrier and actual subcarrier
17:39:18pamauryI read somewhere that the spec says that the 57Khz RDs subcarrier must be phase-locked on the pilot at 19Khz (makes sense form the diagram) but some (many?) emitters do not follow the spec
17:40:13pamaurythus if the tuner recover the pilot at 19Khz and uses that as the source of the 57Khz subcarrier, instead of recovering the 57Khz carrier directly, I expect that you run into some issues
17:58:17 Quit elensil (Quit: Leaving.)
18:17:19 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:24:09 Join paulk-collins [0] (
18:26:15 Quit xorly (Ping timeout: 248 seconds)
18:37:27 Quit MrZeus (Ping timeout: 248 seconds)
19:11:07 Join lebellium [0] (
19:14:08 Join petur [0] (~petur@rockbox/developer/petur)
19:15:06 Quit pamaury (Remote host closed the connection)
19:20:45***Saving seen data "./dancer.seen"
19:34:00 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
19:51:39 Join krabador [0] (~krabador@unaffiliated/krabador)
20:03:12 Quit michaelni (Read error: Connection reset by peer)
20:03:42 Join michaelni [0] (
20:18:18 Quit Moarc (Ping timeout: 272 seconds)
20:18:45 Join Moarc [0] (
20:21:25 Join cereal_killer [0] (d5a244e9@gateway/web/freenode/ip.
20:26:55cereal_killerpamaury: finished successfully and the Makefile was created
20:27:07 Join Bray90820 [0] (
20:27:33cereal_killerBut after typing "make" I get "error: lang.h: No such file or directory"
20:27:38 Quit Bray90820 (Read error: Connection reset by peer)
20:28:15 Join Bray90820 [0] (
20:28:20cereal_killerrule for target ../apps/action.o failed
20:28:33pamaurycereal_killer: try "make clean" and then make
20:29:11 Quit Bray90820 (Read error: Connection reset by peer)
20:34:29 Quit krabador (Quit: Leaving)
20:59:16 Join Bray90820 [0] (
21:00:18cereal_killerpamaury: thanks a lot. with your help, I successfully compiled my first Rockbox build
21:00:26cereal_killernow lets see, what it does on my D2
21:02:42 Join krnlyng_ [0] (
21:05:02 Quit krnlyng (Ping timeout: 240 seconds)
21:11:43cereal_killergevaerts: I compiled a build with the changes you mentioned
21:14:10cereal_killerplayback works fine on the D2, also saving sound or theme settings, but trying to create a bookmark for the currently playing track results in "*PANIC* Updating size on empty dir entry 26"
21:16:55 Nick krnlyng_ is now known as krnlyng (
21:20:46***Saving seen data "./dancer.seen"
21:21:53 Join girafe [0] (
21:22:25cereal_killershould I try the builds before and after the filesystem rework? would this gather more information?
21:22:59pamaurysaratoga: I will promote some targets to stable. I will update the main website, is that ok? (ie not waiting for the release)
21:23:23pamaurycereal_killer: I did not follow closely but I think the consensus was that before the filesystem rework, it worked by pure luck and after it didn't work
21:24:26cereal_killeroh, I see.
21:26:14pamaurybut I might be wrong
21:26:36pamauryanyway, the reworked changed so many things that information before it will probably irrelevant
21:29:06 Quit robertd (Quit: Page closed)
21:35:33 Quit Moarc (Ping timeout: 260 seconds)
21:37:52 Join Moarc [0] (
21:39:07cereal_killerok, thanks for your help.
21:40:11fs-bluebotBuild Server message: New build round started. Revision c534be0, 255 builds, 16 clients.
21:45:11 Quit smoke_fumus (Quit: KVIrc 4.2.0 Equilibrium
21:45:40 Quit cereal_killer (Ping timeout: 240 seconds)
21:48:10 Quit saratoga (Ping timeout: 240 seconds)
21:51:47fs-bluebotBuild Server message: Build round completed after 696 seconds.
21:51:48fs-bluebotBuild Server message: Revision c534be0 result: All green
22:07:40 Join edhelas [0] (~edhelas@
22:13:57 Join saratoga [0] (123e11e0@gateway/web/freenode/ip.
22:17:52saratogapamaury: can you also add a note under "Project news" saying you promoted some devices?
22:25:49pamaurysaratoga: should it be a feature ?
22:30:13 Quit petur (Remote host closed the connection)
22:32:24saratogapamaury: I mean on the front page
22:32:29saratogaabove "3.13 released"
22:33:35lebelliumisn't user manual required to get stable?
22:33:39pamauryah, how does that work? it is hardcoded in the index.t?
22:34:01 Quit Kruppt (Quit: Leaving)
22:36:19saratogalebellium: no, we never required it before, just a wiki page with instructions
22:39:39pamauryI need to update this page too
22:44:26lebelliumsaratoga: oh ok
22:44:42 Quit denniskombucha (Remote host closed the connection)
22:57:21pamauryis it really possible to use A-B repeat on most players?
22:57:31 Quit girafe (Read error: Connection reset by peer)
22:57:36pamauryas far as I can see, it requires dedicated keys which I guess most players cannot spare
22:58:34pamauryor is there another way to set A and B points?
23:03:35 Quit atsampson (Quit: new irssi time)
23:03:56 Join atsampson [0] (
23:10:48lebelliumpamaury: it may be a key combination
23:11:01lebelliumlike power + left and power + right on Clip Zip
23:11:17lebelliumso the lack of keys shouldn't be a huge issue
23:14:28pamaurynot all targets can do combos
23:14:39pamauryand some combos are just crazy impossible (physically)
23:15:39lebelliumwhy can't all targets do combos?
23:16:03pamaurydamn, why does ACTION_STD_MENU means "go to main menu" in all screens except possibly where it means "context menu", which already has an ACTION_STD_CONTEXT...
23:16:16pamaurybecause some target cannot have two button pressed at the same time
23:19:06lebelliumare you referring to some Creative and Sony targets?
23:19:18 Quit ArneB (Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number)
23:19:18 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
23:19:44pamaurylebellium: yes, but I am sure many other targets have this problem
23:20:06pamauryor only some combos, that are definable but I defy anyone to actually use them, are define
23:20:18lebelliumI thought most targets used a combo to boot the OF for example
23:20:32 Join ps-auxw [0] (
23:20:50***Saving seen data "./dancer.seen"
23:21:04 Quit Guinness (Read error: Connection reset by peer)
23:22:19pamauryusually the power button can be combo'ed with everything (but again there are counter example). But the power button make a terrible combo button on most targets because of its location
23:22:20 Quit advcomp2019__ (Ping timeout: 255 seconds)
23:33:41 Quit lebellium (Quit: ChatZilla 0.9.92 [Firefox 49.0/20160916101415])
23:41:47 Join shamus [0] (
23:56:13 Quit paulk-collins (Quit: Leaving)
23:56:14 Quit alexweissman (Read error: Connection reset by peer)
23:56:45 Join alexweissman [0] (

Previous day | Next day