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 2012-01-23

00:03:45 Quit domonoky (Read error: Connection reset by peer)
00:22:17 Join captainkwel [0] (
00:27:14 Join beslayed [0] (
00:28:40bertrikmy laptop build client now advertises arm-ypr0-gcc446 support :)
00:30:31kugelthat's nice. but it doesn't seem to get build complains: "Use of uninitialized value in quoted execution (``, qx) at line 557."
00:31:07bertrikand "Use of uninitialized value in regexp compilation at line 558."
00:31:10kugelI get that too
00:37:17 Quit dfkt (Read error: Connection reset by peer)
00:41:29 Quit pamaury (Remote host closed the connection)
00:42:50 Join Scromple [0] (~Simon@
00:43:14 Quit Scromple (Client Quit)
00:53:21captainkwelkugel, how do you check framerate?
00:53:35kugelcaptainkwel: test_fps.rock
00:53:43captainkwelk, thanks
00:54:46 Join perrikwp [0] (
00:55:43kugelcaptainkwel: you're going to need this patch:
00:56:10kugelthen choose (A)dvanced and then (T)est plugins in configure
00:56:34kugeltest_fps doesn't build or work on touchscreen with current svn
00:56:41kugelgrr, current git
01:07:20***Saving seen data "./dancer.seen"
01:09:21CIA-81Commit 8a43603 in rockbox by Thomas Martitz: Fix flag for the viewport. Text is supposed to be centered.
01:10:39 Quit bertrik (Quit: And That, My Liege, Is How We Know the Earth to Be Banana Shaped)
01:11:06 Quit JdGord (Quit: Bye)
01:11:43CIA-818a43603 build result: All green
01:26:54 Quit Guest28733 (Quit: Konversation terminated!)
01:32:12captainkwelhmm. I'm getting 1/1: 32, 1/4: 61 main , 1/1: 16, 1/4: 47 yuv. hybrid
01:32:14 Quit [Saint] (Read error: Connection reset by peer)
01:32:28captainkwel41, 60, 18, 60 fullscreen
01:32:49 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
01:34:21 Join SynrG [0] (
01:37:40captainkwelfullscreen above meaning the binary in the hybrid build run directly from novaterm
01:38:14JdGordon[Saint]: did you try the touchscreen fix?
01:38:53funmancaptainkwel: 16 fps is not a lot
01:39:18 Quit SynrG (Changing host)
01:39:18 Join SynrG [0] (~synrg@debian/developer/synrg)
02:04:05 Quit Xerion (Read error: Connection reset by peer)
02:04:38 Join Xerion [0] (
02:17:09 Quit Unhelpful (Quit: - Chat comfortably. Anywhere.)
02:17:36 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
02:41:41 Join linuxguy3 [0] (
02:51:31 Quit linuxguy3 (Quit: leaving)
02:52:33 Join linuxguy3 [0] (
03:07:21***Saving seen data "./dancer.seen"
03:53:08 Join Keripo [0] (
04:02:51 Quit Keripo (Quit: Leaving.)
04:05:29 Quit Rob2222 (Read error: Connection reset by peer)
04:05:44 Join Rob2222 [0] (
04:13:13 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
04:16:58 Quit [Saint] (Ping timeout: 252 seconds)
04:19:40 Nick [Saint_] is now known as [Saint] (~Saint]@unaffiliated/saint/x-8516940)
04:22:21 Join mmtt22 [0] (
04:24:24 Quit mmtt22 (Client Quit)
04:37:23 Join Topy [0] (
04:40:43 Quit T44 (Ping timeout: 240 seconds)
04:44:16 Quit ender| (Ping timeout: 276 seconds)
04:45:36 Quit anewuser (Read error: Connection reset by peer)
04:45:38 Join ender| [0] (~ender1@2a01:260:4094:1:42:42:42:42)
04:49:29 Join dys` [0] (
04:51:04 Quit dys (Ping timeout: 252 seconds)
04:52:12 Quit TheSeven (Disconnected by services)
04:52:50 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
04:54:49 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
04:54:58 Quit amiconn (Disconnected by services)
04:54:59 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:55:21 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:55:39 Quit pixelma (Disconnected by services)
04:55:42 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:55:44 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:57:07 Quit [Saint] (Ping timeout: 240 seconds)
05:03:14 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
05:04:03 Join Rob2223 [0] (
05:05:55 Quit [Saint_] (Ping timeout: 252 seconds)
05:07:22***Saving seen data "./dancer.seen"
05:08:18 Quit Rob2222 (Ping timeout: 276 seconds)
05:18:38 Join Amqui [0] (
05:18:39[Saint]I wish we had the ability to check for next AA.
05:18:57[Saint]Even if it wasn't exposed as a tag for skin use.
05:19:14[Saint]Just to stop the horrible redraw on track change.
05:19:24[Saint]Its bugging me. Bad.
05:20:10 Quit Amqui1 (Ping timeout: 260 seconds)
05:20:59[Saint]I can't be a dick about it I guess...the amount of tags that exist basically because "I wanted them" is pretty astonishing. )
05:25:15JdGordonwell noone else is asking for additions so go for it
05:25:21JdGordonI thought next aa could be checked for?
05:25:42JdGordonim not sure how it would work though or if it would fix your problem
05:26:59 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
05:29:34 Quit [Saint] (Ping timeout: 252 seconds)
05:35:23 Join Keripo [0] (
05:39:24 Quit Keripo (Read error: Connection reset by peer)
05:40:57 Join Keripo [0] (
05:49:10 Quit Keripo (Read error: Connection reset by peer)
05:50:38 Join Keripo [0] (
05:53:25 Quit Keripo (Read error: Connection reset by peer)
05:55:03 Join Keripo [0] (
06:09:39 Nick Topy is now known as Topy44 (
06:16:24*[Saint_] has a dilemma
06:16:59[Saint_]I want a generic RaaA manual. However, I hate TeX with a passion...
06:17:16 Join dys`` [0] (
06:17:58 Quit dys` (Ping timeout: 252 seconds)
06:18:07[Saint_]Is there any portions I could write that wouldn't involve me wanting to stab my own face after 10 minutes of TeX?
06:19:04[Saint_]I suspect plain text would be pretty bloody useless as whoever TeXified it would need to do a lot of work.
06:19:49JdGordonIs there really much device dependant TeX that needs doing?
06:20:09JdGordonor better, is there much non-TeX stuff that needs adding for a new manual?
06:20:35JdGordonisnt it just a device picture and fiddl with some config{} thingies?
06:21:33[Saint_]Hmmm...actually, you're probably right. There's a few non-TeX parts that probably wouldn't make any sense at all and news a slight re-word.
06:22:04JdGordonadding the build system changes is probably a good start
06:22:12*JdGordon is 3/4 asleep so probably not making much sense
06:22:14[Saint_]I just want a manual...and I HATE TeX...I was hoping there'd be more I could do.
06:23:45[Saint_]I was hoping "learn TeX" wasn't the answer to contribution here :)
06:25:20 Quit curtism (Quit: Live Long and Prosper)
06:25:48[Saint_]I guess I'll concentrate on doing the thongs I can do. Like finally unifying the various selection fields in Rockbox.
06:25:58[Saint_]*things, lol
06:26:22JdGordonthe what?
06:27:31[Saint_]For instance some settings offer values: 0, 5, 10, 15 etc where others are: 0, 1, 2, 3...for no good reason.
06:27:59[Saint_]In a lot of those situations finer control would be useful.
06:28:28JdGordonI doubt that, but the change should be free
06:30:47[Saint_]Some are actually backward compared to others, also.
06:31:15[Saint_]As in the values increase or decrease in different directions.
06:33:29[Saint_]All I could do would be unify them. A generic function would be much nicer then a tonne of duplication.
06:33:48[Saint_]But that's above MT head.
06:34:11JdGordoni suspect you're going about it wrongly
06:34:42JdGordongo back to your cabbie work... and bug me about the settings thing tonight (if I'm around)
06:36:14[Saint_]The only way I could go about it that I know I won't fuck it up would be to just adjust what's already in place, but tweak the values offered. Which isnt really much of a fix, but more aestheticly pleasing.
06:36:54JdGordonyeah, if you get me when I'm able to look at the code i'll tell you the proper fix
06:37:01JdGordonit should be relativly simple
06:37:54[Saint_]A generic "value between 0 and foo" function that all these could call in its place is what I suspect to be the "correct" fix, but a bit over my head.
06:38:10JdGordonthats already there
06:38:20JdGordoni promise if you can handle skin code you can handle settings_list.c
06:39:02JdGordonand as you're apparently a bit OCD (don't mean that as an offence) there is plenty of reordering/cleanup worth doing in that file :)
06:39:16JdGordonput all the related settings together in the config instead of mostly randomly
06:40:12[Saint_]I am, yes. I don't take it as a bad thing, and the bonus is its usually the tiny things no one else can be bothered doing :)
06:40:46[Saint_]I just don't like differences for no good/immediately apparent reason.
06:41:07JdGordonthere used to be a good reason.... err... 4+ years ago
06:41:24JdGordonthen i did a straight system replace to how the configs were loaded and that reaosn vanished
07:07:27***Saving seen data "./dancer.seen"
07:07:38 Quit captainkwel (Ping timeout: 245 seconds)
07:16:49 Join PaulJam [0] (
07:31:43 Quit PaulJam (Ping timeout: 240 seconds)
07:42:49 Quit [Saint_] (Remote host closed the connection)
08:08:27 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
08:09:52 Join GodEater_ [0] (93722cd1@rockbox/staff/GodEater)
08:29:44 Quit [Saint] (Quit: Quit)
08:38:04 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:58:01 Quit Keripo (Quit: Leaving.)
08:58:38kugelzagor: i added the ypr0 to the builds file but it wasn't build by the build system
09:07:10 Quit Zarggg (Ping timeout: 252 seconds)
09:07:30***Saving seen data "./dancer.seen"
09:11:59Zagorah, the build master still does "svn up builds"...
09:16:35 Join einhirn [0] (
09:20:58 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
09:24:04kugelis the build table now sorted correctly?
09:24:24Zagoryes, just fixed that
09:27:37 Join nosa [0] (
09:28:30amiconnAre the deltas calculated between the correct revisions? The table looks rather weird to me
09:29:07 Quit nosa-j (Ping timeout: 244 seconds)
09:29:07 Nick nosa is now known as nosa-j (
09:29:24Zagoramiconn: I'm not sure. I agree it looks strange.
09:32:01Zagorthis looks more sane
09:32:48ZagorI'm still not sure though :-)
09:33:08amiconnStill not correct though
09:33:41amiconnCheck e.g. 490c87c. It's just a change to .gitignore, so shouldn't cause any delta (except the occasional +/-16)
09:35:12 Join wodz [0] (
10:02:45 Join efyx [0] (
10:03:59JdGordon[Saint]: ok, whats the issue with settings? which ones specifically?
10:08:45 Join pamaury [0] (
10:08:45 Quit pamaury (Changing host)
10:08:45 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:09:30[Saint]Ones I've notices are: Settings - Playback Settings - Rewind Before Resume, * - * - Rewind On Pause, and * - * - Skip Length. The first two specify different units (one in multiples of one, the other multiples of five), the latter runs backwards compared to the former two.
10:09:40[Saint]There's likely other instances.
10:10:12Zagoramiconn: actually, it is correct (but confusing). there were a dozen commits between 109084d and 490c87c.
10:11:06Zagorin a commit burst (as we're going to have a lot more with git than with svn), we don't build every single commit.
10:11:47[Saint]...sounds odd.
10:12:24JdGordonwe should stop doing burst commits...
10:12:25Zagorit becomes more confusing with the hashes since it is not visible there were commits inbetween
10:12:46[Saint]JdGordon: beat me to it :)
10:13:30ukleinekZagor: you can log the number of commits between the last build and the recent one.
10:14:37Zagoryes, we've been discussing using a "lastrelease+count" commit naming scheme on the web pages
10:15:11[Saint]Note to all: If anyone else is aware of any instances in the settings where the values aren't multiples of one and it doesn't make any sense for them not to be, can they make a note of it here please?
10:21:02[Saint]Teehee...actually, the examples I listed above all list different units. One in steps of 1, one in steps of 5, and one in primes until 10 and then steps of 5 :-S
10:21:12JdGordon[Saint]: there is a magic flag settings need to make the list invert if it is backwards
10:22:34[Saint]Skip Length is amusing...1, 3, 5, 7 (misses 9), 10, 15...etc.
10:23:09kugelit's kind of logarithmic scale
10:23:31kugeltiny steps don't make sense if the values get bigger IMO
10:23:59[Saint]0-240 should cut it for all. And give finer control at (probably?) no cost.
10:24:37kugelbeing a longer list is a cost on its own
10:25:03ukleinekZagor: but that will only work if you release from the master branch or merge the release branch back into master. Otherwise you won't have tags reachable from master.
10:25:11kugelskip length is a nice list if you ask me. the steps gradually increase with the values
10:25:30gevaerts[Saint]: having to scroll through hundreds of items is no cost?
10:26:54[Saint]Yes. I just thought of that. Its the lower values I want finer control over personally. There's been a few times where I've wanted to select say 2 seconds, and can only select 0 or 5.
10:27:47kugelre: backwards, i think some lists are "backwards" depending if built on a scroll wheel target
10:28:18[Saint]0-10 in units of 1, 10-60 in units of 5, and units of 10 from then on would probably do.
10:28:19kugelso that scrolling forward increases values (on other targets the biggest values are at the top)
10:30:14kugelI think it's strange that we have rewind on pause and rewind on resume individually
10:31:10JdGordoni tihnk its strange people think its strange we have settings bloat
10:31:13Zagorukleinek: git log −−online v3.10-final..HEAD | wc -l
10:31:16[Saint]kugel: re: backward settings, skip length and rewind on pause differ in their "orientation" on a touch target. I know about the scroll wheel thing, but I found this odd.
10:32:55 Join TomColler [0] (
10:33:17[Saint]Hmmm...skip length is an odd creature. It goes up to 15 minutes.
10:33:30ukleinekZagor: right, that works, but git describe will fail.
10:33:56ukleinekand the number isn't very helpful in general
10:34:11Zagorno, only to illustrate commit sequence
10:35:08[Saint]Hmmmm...0-10 in units of 1, 10-60 in units of 5, and units of 60 (presented as one minute, of course) there on?
10:35:36*ukleinek takes back the comment about helpfulness. It's just in another way unhelpful compared to the numbers of git-describe.
10:35:39JdGordon[Saint]: settings_list.c is to become your nemisis....
10:36:04JdGordon"scan min step" being backwards is nice and easy to fix
10:36:27JdGordonerr, well, its correct on wheel targets as it is now though
10:39:00[Saint]I've found there's no "one sane way" to do this for all...but, the values should a: have the same steps, and b: run in the same direction (IMO).
10:39:19JdGordonb is agreeable
10:39:28JdGordona really needs a better setting chooser ui
10:39:39*[Saint] nods
10:40:19JdGordonpretty much the steps for every setting is arbitrary... if there was a way to change the step size (based on scroll speed maybe?) then we would only need min and max and yeah
11:07:32***Saving seen data "./dancer.seen"
11:20:38 Join leavittx [0] (~leavittx@
11:21:53 Quit perrikwp (Read error: Connection reset by peer)
11:23:11 Join perrikwp [0] (
11:28:48 Join stoffel [0] (
11:33:54 Quit leavittx (Ping timeout: 240 seconds)
12:04:14 Quit nosa-j (Ping timeout: 240 seconds)
12:20:44 Quit stoffel (Ping timeout: 260 seconds)
12:23:02 Quit perrikwp (Read error: Connection reset by peer)
12:24:24 Join perrikwp [0] (
12:25:39 Join nosa-j [0] (
12:26:09 Quit perrikwp (Read error: Connection reset by peer)
12:27:20 Join perrikwp [0] (
13:02:34JdGordon[Saint]: you putz! have you tried the touchscreen fix yet?
13:02:55*JdGordon toiled on that fix for *literally 15min* and what do I get?
13:02:58JdGordonnot even a thank you!
13:05:04[Saint], I actually haven't. I've not found the time to set up a new dev environment as yet. I was going to yesterday but it took far too long to rebuild the raid array.
13:07:34***Saving seen data "./dancer.seen"
13:17:15 Quit wodz (Quit: Leaving)
13:20:58[Saint]Is the build farm building RaaA yet?
13:21:15[Saint]Can't say I've noticed if it is.
13:21:36gevaertsIt is, yes
13:21:49[Saint]If it isn't, is there a reason? (Signing?)
13:22:14gevaertsBut only building, not making available :)
13:22:28[Saint]gevaerts: do you know how signing is being handled?
13:22:37gevaertsIt isn't
13:22:50[Saint]Or is it not, since there's no distribution.
13:23:21gevaertsWe build RaaA for SDL and Android
13:23:32 Quit kugel (Ping timeout: 252 seconds)
13:26:23[Saint]I can only think of three ways to handle signing, and I'm not sure about any of them. 1: a single trusted machine, 2: building from the farm and passing back to a single trusted machine for signing, 3: sharing a single key amongst the entire farm (possibly error prone?)
13:26:43gevaerts(3) is not an option
13:27:03[Saint]...its not?
13:27:15gevaertsWe do want *some* sort of security
13:28:01gevaertsSharing a private key amongst dozens of machines is madness
13:29:03[Saint]Its SPARTA!
13:30:26JdGordonespecially if we are actually oging to add network stuff into RaaA
13:31:56[Saint] does assume we don't have assholes deliberately poisoning builds.
13:32:17gevaertsIt also assumes that those machines are all equally secure
13:33:11gevaertsWith central signing, if one build machine is compromised, we may have a few bad builds, but we can easily recover. With the key everywhere, it means we have to get a new key, which has all sorts of nasty consequences
13:34:56[Saint]So unless I'm mistaken, of the options I thought of (unless there's something I missed or I'm not aware of?) the "single trusted machine" is probably the best option?
13:34:57JdGordonis there any reaosn we cant have the build master (which is a client as well right?) doing the android signed build always?
13:35:27gevaertsJdGordon: it's not a client, no
13:35:34 Join stoffel [0] (
13:35:47gevaertsIt should be able to do the signing after uploading though
13:36:46[Saint]Oh...derp! I failed to account for builds being passed back to the build master anyway. Whoops.
13:36:52gevaertsAnd if we want a bit more security, uploads could be signed separately so we know they come from the right machine. That would be an internal thing though, not something visible from the outside
13:38:44 Quit SynrG (Ping timeout: 248 seconds)
13:39:43[Saint]RaaA builds being made available would be awesome. A lot of people don't seem to trust the rasher builds.
13:41:08 Part TomColler
13:43:02gevaertsRasher's builds should be just as trustworthy as buildsystem-provided builds would be
13:48:41[Saint]But there's the huge "not affiliated in any way" warning...and they're available from the unsupported builds section.
13:49:13[Saint]People probably have no idea who rasher is...
13:51:09gevaertsWell, there's the "not official or sanctioned" warning, which I think isn't the same as "not affiliated"
13:52:18[Saint]Ah, whoops...I remembered incorrectly.
13:52:45JdGordonthe way we are doing android builds is just wrong anyway
13:52:54gevaertsIt's true that he doesn't say that he's not not affiliated :)
13:53:03gevaertsJdGordon: in what way?
13:54:25 Nick perrikwp is now known as perrikwp_ (
13:54:26 Nick perrikwp_ is now known as perrikwp (
13:55:47JdGordonin that the apk contains the file... the apk should just be enough to almost bootstrap that, which can then be updated over the net from the java bits
13:56:28gevaertsAh, right. I don't really have an opinion on that
13:57:38*[Saint] is shocked
14:05:00Tornei would suggest the opposite, personally
14:05:06Torneteh apk should be complete and functional
14:05:36Tornerockbox works offline on other devices, why shouldn't it on android?
14:06:28Tornethe more android-y we become the lower the probability that updates can be done without changing the java code anyway
14:06:42Tornei don't think trying to roll in our own update mechanism there is going to gain us anything
14:07:08Torneit'll just mean that if you install rockbox from teh market, then don't launch it until later when you don't have a network connection any more, it won't work :)
14:07:20 Join lorenzo92 [0] (
14:09:56CIA-81Commit 0152541 in rockbox by Thomas Martitz: gitignore: Change to generic .so.
14:10:14 Quit Amqui (Ping timeout: 240 seconds)
14:11:27lorenzo92kugel: I've seen you're working to fix all the warnings to be compliant with the auto-builder ;)
14:12:17lorenzo92One thing before I go, I think you can merge the gpio buttons patch. I tested it for at least 15 days and worked fine so far...
14:12:23CIA-810152541 build result: 1 errors, 0 warnings (Thomas Martitz committed)
14:13:10 Quit dionoea (Ping timeout: 252 seconds)
14:13:45gevaertsZagor: that 1 error seems wrong
14:15:41 Join Amqui [0] (
14:16:10 Join kugel [0] (~kugel@
14:16:11 Quit kugel (Changing host)
14:16:11 Join kugel [0] (~kugel@rockbox/developer/kugel)
14:16:36kugelman, still no ypr0 in the build system
14:18:19 Join leavittx [0] (~leavittx@
14:18:38kugellorenzo92: oh, I thought it was merged already
14:18:59lorenzo92kugel: no problem ;)
14:23:44 Quit kugel (Quit: leaving)
14:25:21 Quit lorenzo92 (Quit: ChatZilla 0.9.88 [Firefox 9.0.1/20111228084940])
14:27:24 Join kugel [0] (~kugel@
14:27:31 Quit kugel (Changing host)
14:27:32 Join kugel [0] (~kugel@rockbox/developer/kugel)
14:29:33 Quit factor (Quit: Leaving)
14:34:05 Quit leavittx (Remote host closed the connection)
14:36:26Zagorkugel: that's because your client disconnected during the round. so "Nobody can build samsungypr0. Disabling target."
14:37:08kugelZagor: hm, m client was disconnected when the build round started
14:37:37kugelbertrik's client can also build it
14:38:34kugelbut perhaps bertrik isn't online
14:38:48kugel(I didn'T know my client isnt connected when I made the commit)
14:46:19kugelZagor: btw, throws this warning Use of uninitialized value in quoted execution (``, qx) at line 557.
14:46:22kugelUse of uninitialized value in regexp compilation at line 558.
14:54:16CIA-81Commit 016bc24 in rockbox by Thomas Martitz: ypr0: Allow dection of multiple button presses at the same time.
14:55:48Zagorkugel: that warning is just because the new compiler hasn't been added to rbclient's compiler check
14:56:04 Join dfkt [0] (dfkt@unaffiliated/dfkt)
14:56:42CIA-81016bc24 build result: All green
14:58:28kugeloops, /me should have rewritten author information for git
15:03:05 Quit nosa-j (Ping timeout: 252 seconds)
15:07:38***Saving seen data "./dancer.seen"
15:08:02 Join nosa-j [0] (
15:08:59Torneanyone have any thoughts on what we want to require before a change can be submitted through gerrit (i.e. by a non-committer)?
15:09:12TorneThe default gerrit review process has "Code-Review" and "Verified" as categories
15:09:39TorneVerified is supposed to mean "someone actually tested that this accomplishes whatever it claims to accomplish (fixes bug, feature works, whatever)
15:10:11Torneyou need a positive in every category in order to submit, and no negatives
15:10:26Tornebut the set of categories is arbitrary
15:10:32Zagorin order to commit?
15:10:45Torneto submit a change via gerrit
15:10:58TorneIf you push to master directly you can push what you like
15:11:33Zagoryeah but in gerrit you have multiple submits: submit to review and submit to git. we should probably not call both submit.
15:12:01Tornegerrit consistently refers to submitting something to the master branch as submit; the button is labelled submit
15:12:09Torneuploading a change for review is.. uploading a change
15:12:35Torneusing "commit" for these things is worse than submit, imo, since with git commit is something you do locally before anything else :)
15:12:41 Join LinusN [0] (
15:12:59Zagoryeah, it's just rather confusing when submitting a patch is not submitting a patch :-)
15:13:20 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
15:13:43Torneanyway, not relevant :@)
15:13:52TorneWe need at least one category, and the code-review default category is sensible
15:14:37Tornethe usual config for that is that project developers can give +2 or -2 code review, and otehr people can give +1 and -1
15:14:53Torneyou need at least one +2 and no -2, in order to get your change incorporated into the branch
15:15:07Tornethe +1 and -1 are advisory only; they don't enable or block commits, but they let other people voice an opinoin clearly
15:15:30Tornethat seems reasonable to me; the questoin is do we care about having a second (or further) categories for things like "this has been tested"
15:15:32Zagorsounds reasonable
15:15:57ZagorI don't think we want that, not to begin with first anyway.
15:18:06Torneyeah. i am inclined to agree. We can always add it later, if we feel it's warranted
15:18:24Tornethe way we use this at work is that usually people set their own commits to +1 Verified, beacuse they have tested them
15:18:43Tornebut if they don't have the ability to test it themselves for whatever reason they can ask someone else to
15:18:59Tornebut most of the time this is just one more button you have to click, so the benefit is not huge :)
15:19:11Zagorsure, having a field for "I have actually verified this on target" could be informative, but should not be required
15:19:30Tornehm, actually
15:19:37Torneit might be possible to do that
15:20:03Tornethere is a way to assert that this category *doesn't* block commits, i.e. a positive is still needed but negatives don't block it
15:20:24Tornemaybe you can disable the policy enforcement and just have it FYI.. *looks*
15:23:09CIA-81Commit 8517593 in www by Björn Stenberg: Fill paragraphs in commit messages.
15:23:39 Quit tchan (Read error: Connection reset by peer)
15:24:25 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
15:25:28TorneYeah, there is.
15:25:50TorneOK, so that might be worth keeping but just changing the type to NoBlock?
15:26:14Tornesuch that it's advisory only
15:26:27Torneprobably also want to rename it from Verified to Tested? :)
15:26:40Zagorsounds good to me
15:26:56Tornealso if it's advisory only we can probably safely let anyone use it
15:27:13Tornesuch that users can just say "i tried this, wfm" or vice versa
15:27:30Torneok. i'll poke the database
15:34:55TorneZagor: can you restart gerrit to pick up the change?
15:35:06Tornethen i'll try it on the sandbox repo
15:38:44 Join SynrG [0] (
15:39:06CIA-81Commit 531fb01 in sandbox by Torne Wuff: another test
15:39:39CIA-81Commit d3d1422 in sandbox by Torne Wuff: One more
15:39:44TorneZagor: OK, that works great.
15:39:59Torneyou can mark things as "works on target" or "fails on target"
15:40:06 Join dionoea [0] (
15:40:11Tornebut it doesn't matter with regards to being allowed to submit it
15:40:23Tornewe can easily change this later, it's pretty easy :)
15:40:42Torneok, so that's one thing. *yay*
15:40:54 Join perrikwp_ [0] (
15:42:49TorneZagor: Hm, one more thing with this, actually: who should be allowed to actually press the submit button (once the constraint of "a committer has reviewed it positively and no committer reviewed it negatively" has been satisfied
15:42:58Tornejust committers? or should we let peple do it themselves?
15:43:10 Join nosa [0] (
15:43:21Zagoronly committers imho
15:43:29Torneok. we can always revisit it later.
15:43:37 Quit perrikwp (Ping timeout: 240 seconds)
15:43:50 Quit nosa-j (Ping timeout: 272 seconds)
15:43:51 Nick nosa is now known as nosa-j (
15:43:53TorneThe main thing there would be "don't commit something for someone if they haven't indicated *they* think it's ready" :)
15:43:59Tornebut hey
15:44:02Tornethat's not a major risk
15:46:26 Join perrikwp [0] (
15:48:41 Quit perrikwp_ (Read error: Operation timed out)
15:48:51Tornei'm gonna leave the submit method being cherry-pick, also
15:49:34Tornepeople seem reasonably happy with the minimise-merges approach to master, so we don't want gerrit to go and make a bunch of merges for us, and fast-forward-only is *incredibly* tedious since you have to rebase your change and reupload it and get it rereviewed if you are not completely up to date
15:49:57 Quit stoffel (Remote host closed the connection)
15:50:14Tornethe downsides of cherrypicking are 1) it doesn't enforce dependencies between patches *yet* but will in a future gerrit version and 2) it means after you submit you need to rebase your local branch to make it "notice" that your commit has been included, since it won;'t have the same hash
15:50:25Tornethe upside is gerrit can insert the lines in the message saying who reviewed it/etc
15:50:36 Join stoffel [0] (
15:51:21 Quit SynrG (Changing host)
15:51:21 Join SynrG [0] (~synrg@debian/developer/synrg)
15:52:08 Quit stoffel (Remote host closed the connection)
15:54:33 Join stoffel [0] (
15:55:52kugelTorne: cherry-picking doesnt change the hash does it?
15:55:57Tornekugel: yes, absolutely
15:56:10Tornethe hash is computed from the patch, the commit message *and* the parent commit
15:56:16Tornecherrypicking changes at minimum the parent commit
15:56:30Torneand gerrit also changes the commit message to add the reviewed-by lines and so on (since it might as well, as the hash has already changed)
15:56:53TorneIf you rebase afterward then the rebase will just delete your local copy since there will be no patch to apply any more
15:57:00Tornebut it doesn't trivially disappear the way it would with a merge.
16:00:02CIA-81Commit a2b9466 in www by Björn Stenberg: Converted build farm to git.
16:02:36TorneWould someone like to review and commit,64 for me? :)
16:02:43Tornejust to test that it's working :p
16:02:53Torne(yes, to the real repo, not the sandbox)
16:03:30 Quit dionoea (Changing host)
16:03:30 Join dionoea [0] (~dionoea@videolan/developer/dionoea)
16:04:36kugelTorne: I don't see the tested and reviewed buttons
16:04:43Tornekugel: are you logged in?
16:04:49CIA-81Commit 2ab2857 in rockbox by Björn Stenberg: Remove svn id.
16:05:11TorneZagor: we probably want CIA to read author, not committer?
16:05:17Torneas that should really have said it was by me :)
16:05:49Zagoror both. I triggered the action, so I'm sort of taking responsibility.
16:06:03Torneyeah, if theya re differnt it might be worth including both
16:06:18TorneSo yeah, that's all working now
16:06:24kugelI don't think the real author is important for irc announcement
16:06:26TorneI will update UsingGit in a mo to explain to people how to use it
16:06:44kugelTorne: yes I am logged in
16:06:45Tornebut it is enabled now, so you can try if you like (see GerritDemoGuide for the outline) :)
16:06:52CIA-812ab2857 build result: All green
16:06:58Tornekugel: Where were you looking?
16:07:05TorneYou need to hit "Review" on the patchset
16:07:14kugelthe page you linked to
16:07:16Tornethen as well as entering review comments you can select your vote for reviewed/tested
16:07:22TorneYeah, they are not direclty buttons there
16:07:28TorneThey're radio buttons inside the review-comment form
16:07:56kugeloh okay. I didn't see the review button
16:08:50kugelare we limited to one patch per ticket?
16:09:11kugel(one patch as in one change including newer versions of it)
16:10:12 Quit perrikwp (Read error: Connection reset by peer)
16:11:23 Join perrikwp [0] (
16:14:27CIA-81Commit 2ab2857 in rockbox by Björn Stenberg: Remove svn id. ...
16:14:39Tornekugel: one *current* patch per change
16:14:58TorneIf you make multiple commits and push them to gerrit together, they will be uploaded as multiple changes with dependencies between them
16:16:13CIA-81Commit 2ab2857 in rockbox by Björn Stenberg: (Author: Torne Wuff) Remove svn id.
16:16:35Tornekugel: try it in the sandbox :)
16:16:47Tornejust commit more than once and push HEAD to refs/for/master
16:16:50Torneand you'll see what you get.
16:17:47Torneand yes, you can add messages/etc post-merge and they get emailed to the relevant people
16:20:44 Quit perrikwp (Read error: Connection reset by peer)
16:21:06ZagorTorne: do you know if gerrit can be configured to use something other than `hostname` as From: host?
16:21:18Tornei expect so :)
16:22:13Torneyeah, set sendemail.from
16:22:18 Join perrikwp [0] (
16:22:44 Quit perrikwp (Read error: Connection reset by peer)
16:23:16ZagorI'll add a gerrit alias to
16:23:29Tornealso and
16:23:41Tornewill determine what it calls itself in git, if it creates commits (e.g. merges)
16:23:52Torneshould probably be set the same :0
16:23:55 Join perrikwp [0] (
16:31:20 Join WalkGood [0] (~4@unaffiliated/walkgood)
16:39:41 Part LinusN
16:42:38 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
16:43:07kugelpZagor: we need an organization email for Github too
16:43:59Zagorkugelp: created
16:45:10kugelpokay, will change it if you don't beat it (i invited your as a org owner)
16:45:25kugelpautomatic mirroring would be nice too
16:46:45 Quit nosa-j (Ping timeout: 255 seconds)
16:50:03kugelpZagor: who receives the mails sent to
16:50:09ZagorI do
16:51:18kugelpyou can add me too if you like, but I guess its not strictly necessary
16:51:41 Join nosa-j [0] (
16:52:18ZagorI don't expect a lot of traffic :)
16:56:08 Part Zagor
17:00:05 Quit tmr (Quit: ZQ)
17:05:21 Quit Galois (Ping timeout: 252 seconds)
17:07:41***Saving seen data "./dancer.seen"
17:10:15 Join PaulJam [0] (
17:10:51TorneI have updated UsingGit to cover uploading changes for review, reviewing changes, and submitting changes once they have been reviewed.
17:11:04TornePeople: have a go at some point, and let me know if the instructions need to be improved.
17:11:42TorneWhere do we stand on the rest of the infrastructure?
17:22:22 Join perrikwp_ [0] (
17:23:42 Quit perrikwp (Read error: Operation timed out)
17:25:42 Quit perrikwp_ (Read error: Connection reset by peer)
17:27:11 Join perrikwp [0] (
17:29:35 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
17:30:07 Quit PaulJam (Ping timeout: 245 seconds)
17:33:23 Quit amithkk (Read error: Connection reset by peer)
17:36:00 Join GeekShad1w [0] (
17:37:06 Join perrikwp_ [0] (
17:37:16 Quit GeekShadow (Ping timeout: 252 seconds)
17:38:01 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:38:42 Quit perrikwp (Quit: No Ping reply in 180 seconds.)
17:46:35 Join amithkk [0] (u4289@2buntu/writers/amithkk)
17:47:54 Quit nosa-j (Ping timeout: 252 seconds)
17:48:18 Join dfkt [0] (dfkt@unaffiliated/dfkt)
17:50:11 Join nosa-j [0] (
17:50:25 Join anewuser [0] (~anewuser@
17:50:26 Quit anewuser (Changing host)
17:50:26 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
17:50:36 Nick Jack87|Away is now known as Jack87 (Jack87@nasadmin/admin/jack87)
17:52:49 Nick Jack87 is now known as Jack87|Away (Jack87@nasadmin/admin/jack87)
17:58:15 Join y4n [0] (y4n@unaffiliated/y4ndexx)
18:14:40 Quit GeekShad1w (Quit: leaving)
18:19:34 Quit kugelp (Remote host closed the connection)
18:19:40 Join kugel [0] (~kugel@
18:19:40 Quit kugel (Changing host)
18:19:40 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:19:43 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
18:20:16 Join remlap1 [0] (
18:20:27 Quit remlap1 (Client Quit)
18:20:44 Join lorenzo92 [0] (
18:22:02 Quit remlap (Ping timeout: 240 seconds)
18:22:20 Join GeekShadow [0] (
18:25:46 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:32:32 Quit lorenzo92 (Quit: ChatZilla 0.9.88 [Firefox 9.0.1/20111228084940])
18:37:12 Quit GeekShadow (Quit: leaving)
18:37:53 Join GeekShadow [0] (
18:41:08 Quit GeekShadow (Client Quit)
18:43:00 Join GeekShadow [0] (
18:43:03 Join perrikwp [0] (
18:45:02 Quit perrikwp_ (Read error: Operation timed out)
18:45:25 Join remlap [0] (
18:49:21 Quit GeekShadow (Quit: leaving)
18:51:15 Join y_square [0] (
18:51:19 Join GeekShadow [0] (~antoine@
18:53:11 Join Strife89 [0] (~Strife89@
19:04:23 Join lebellium [0] (
19:05:30 Join Galois [0] (
19:05:49 Quit perrikwp (Read error: Connection reset by peer)
19:07:02 Join perrikwp [0] (
19:07:43***Saving seen data "./dancer.seen"
19:20:51 Join PaulJam [0] (
19:24:39 Quit stoffel (Remote host closed the connection)
19:27:24 Quit benedikt93 (Quit: Bye ;))
19:31:48 Quit Strife89 (Ping timeout: 260 seconds)
19:35:55 Join n1s [0] (
19:35:55 Quit n1s (Changing host)
19:35:55 Join n1s [0] (~n1s@rockbox/developer/n1s)
19:38:12 Join perrikwp_ [0] (
19:40:02 Quit perrikwp (Ping timeout: 240 seconds)
19:45:48 Quit Torne (Ping timeout: 276 seconds)
19:49:18 Quit WalkGood ()
19:51:13 Quit n1s (Read error: Operation timed out)
19:53:16 Join Torne [0] (~torne@rockbox/developer/Torne)
19:54:06 Quit mc2739 (Ping timeout: 240 seconds)
19:55:28 Quit perrikwp_ (Read error: Connection reset by peer)
19:56:39 Join perrikwp [0] (
19:58:26 Quit Torne (Ping timeout: 272 seconds)
20:07:16 Join TheLemonMan [0] (
20:11:55 Quit perrikwp (Read error: Connection reset by peer)
20:13:08 Join perrikwp [0] (
20:17:21 Join n1s [0] (
20:17:31 Quit n1s (Changing host)
20:17:31 Join n1s [0] (~n1s@rockbox/developer/n1s)
20:21:51 Join bertrik [0] (
20:21:51 Quit bertrik (Changing host)
20:21:51 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
20:24:16 Join perrikwp_ [0] (
20:26:38 Quit PaulJam (Ping timeout: 260 seconds)
20:26:47 Quit perrikwp (Ping timeout: 245 seconds)
20:39:29 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
20:45:02 Quit perrikwp_ (Read error: Connection reset by peer)
20:46:13 Join perrikwp [0] (
20:55:32 Quit funman (Ping timeout: 248 seconds)
20:57:17 Join funman [0] (~fun@rockbox/developer/funman)
21:01:15 Join liar [0] (
21:05:09 Quit funman (Ping timeout: 252 seconds)
21:05:45 Join funman [0] (~fun@rockbox/developer/funman)
21:07:45***Saving seen data "./dancer.seen"
21:13:33desowinJdGordon: (quite late, been busy) Sansa Connect port does not have RTC support, as I have no idea if the RTC is even connected to DM320. OF doesn't seem to use it, so the bug you mentioned (FS #12247) does not apply here
21:13:34fs-bluebot Can't set time of day clock on mrobe mr500i (bugs, unconfirmed)
21:14:14 Quit n1s (Ping timeout: 240 seconds)
21:15:03desowinhmm, 29234.. if that was later, that could mean some udelay() is too low, as I have fixed it to be precise
21:15:21 Quit beslayed (Remote host closed the connection)
21:15:51desowin(Connect was commited in 31000)
21:16:27 Quit bluebrother^ (Read error: Connection reset by peer)
21:16:31 Quit fs-bluebot (Read error: Connection reset by peer)
21:19:47 Join fml [0] (
21:20:48fmlWhat is the role of FlySpray now that we have gerrit? Are people still supposed to submit patches on FlySpray?
21:21:13 Join curtism [0] (
21:21:31 Join bluebrother [0] (
21:21:31 Quit bluebrother (Changing host)
21:21:31 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
21:21:37 Join fs-bluebot [0] (
21:26:20fmlIMO the point above should be covered on the UsingGit wiki page
21:26:25fmltorne ^^
21:37:50 Quit TheLemonMan (Quit: WeeChat 0.3.6)
21:38:51 Quit fml (Quit: ChatZilla 0.9.88 [Firefox 9.0.1/20111220165912])
21:40:21 Join perrikwp_ [0] (
21:40:55 Join Administrador [0] (
21:41:16Administradorhow can i choose the folders i want rockbox to scan+
21:41:37bertrikby default it scans everything
21:41:42Administradorits showing me audio files for games
21:41:59Administradoris there any way to change that?
21:43:18 Quit perrikwp (Ping timeout: 255 seconds)
21:43:45bertrikI don't know, but I guess it's in the user manual
21:44:26gevaertslook for database.ignore
21:45:45gevaertsin the manual
21:50:04Administradorcant find it
21:51:13 Join MethoS- [0] (~clemens@
21:58:40Administradoranyone help+?
22:02:13 Join Strife89 [0] (~Strife89@
22:03:35alexbobpyou can't find the manual?
22:03:43alexbobpwhat did you download rockbox from, some torrent site?
22:03:46alexbobprockbox has a website you know :P
22:03:56Administradori found the manual
22:04:15Administradorbut cant see where it says how to select the folders i want to listen to
22:04:19Administradorin database
22:04:35 Quit y_square (Quit: CGI:IRC (Ping timeout))
22:06:01alexbobpit's in the manual. search for exactly what gevaerts told you to search for.
22:06:19Administradori did
22:07:09alexbobpI don't believe you
22:07:13alexbobpbecause I searched and it's there
22:07:23alexbobpin 4.2.2
22:07:25Administradorfound it
22:07:51Administradordo i have to do it from the computer?
22:07:55alexbobpif you spent as much time actually looking as you spent asking repeatedly on irc, you would have found it three times and gotten laid by now :D
22:08:11alexbobpit just says to create a file with that name. why would it care what you make the file with>?
22:09:37 Join AlexP [0] (~alex@rockbox/staff/AlexP)
22:09:55Administradorhow can i make the battery last longer without sacrificing sound quality?
22:11:37 Quit perrikwp_ (Read error: Connection reset by peer)
22:11:57 Join n1s [0] (
22:11:57 Quit n1s (Changing host)
22:11:57 Join n1s [0] (~n1s@rockbox/developer/n1s)
22:12:56 Join perrikwp [0] (
22:12:58alexbobpjust click on "shopping" ;)
22:16:29 Quit Administrador (Quit: ChatZilla 0.9.87 [SeaMonkey 2.2/20110706120824])
22:21:04 Join perrikwp_ [0] (
22:23:14 Quit perrikwp (Ping timeout: 240 seconds)
22:24:26 Join perrikwp [0] (
22:26:42 Quit perrikwp_ (Ping timeout: 240 seconds)
22:30:00[Saint]alexbobp: a little harsh, dude.
22:30:30[Saint]I agree, people should search...however, Leta try not to abuse people huh? :)
22:33:48alexbobpsorry... we was just demonstrating such an amazing lack of effort :/
22:34:26[Saint]Lol...nice typo :)
22:35:28 Quit AlexP (*.net *.split)
22:35:28 Quit ThomasAH (*.net *.split)
22:35:45 Join AlexP [0] (~alex@rockbox/staff/AlexP)
22:37:18 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
22:40:37 Join ThomasAH [0] (
22:40:48 Quit [Saint] (Ping timeout: 248 seconds)
22:41:51 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
22:42:28 Join beslayed [0] (
22:42:36 Quit [Saint_] (Ping timeout: 240 seconds)
22:43:37 Quit curtism (Quit: Live Long and Prosper)
22:44:07 Join curtism [0] (
22:44:32beslayedhas anyone had trouble with a rockboxed sansa clip zip getting stuck in the "refreshing your media" screen?
22:45:40[Saint]Only when the filesystem is trashed.
22:45:54beslayedyes, i think that's what happened
22:46:45beslayedok, i finally got into the of screen to reformat...
22:46:59[Saint]An unclean shutdown, or a whole host of other fun things can cause it. Luke failing to properly safely eject/unmount.
22:49:39[Saint]Filesystem would be the first thing I'd check, but, if that's a bit of a bother for <some reason> then formatting will have the same effect...its just a bit more drastic.
22:51:20 Quit perrikwp (Read error: Connection reset by peer)
22:51:33 Quit n1s (Quit: Ex-Chat)
22:52:32 Join perrikwp [0] (
23:01:28dfkti want to use the x5's power slider to turn the screen off on the wps screen (in other screens it's mostly used for escape) - is there some action predefined for this? - { ACTION_STD_?????, BUTTON_POWER, BUTTON_NONE },
23:03:55beslayed[Saint] yeah, I tried checking the filesystem, but couldn't get it fixed. Finally managed to get into the of and reformat. Re-rockboxed, and working fine now.
23:07:41 Join perrikwp_ [0] (
23:07:49***Saving seen data "./dancer.seen"
23:10:05 Quit perrikwp (Ping timeout: 244 seconds)
23:10:15JdGordondfkt: no
23:11:54dfktJdGordon, i assume no other target has 'voluntary' screen-off as well?
23:20:30 Join Torne [0] (~torne@rockbox/developer/Torne)
23:27:00 Join Thra11 [0] (~thrall@
23:27:04 Join perrikwp [0] (
23:27:06 Quit perrikwp_ (Read error: Operation timed out)
23:29:44 Quit MethoS- (Quit: Konversation terminated!)
23:30:19 Quit liar (Remote host closed the connection)
23:32:02 Quit bertrik (Ping timeout: 240 seconds)
23:38:39 Join lovasoa [0] (~yaaic@
23:39:45 Join lovasoa2 [0] (~yaaic@
23:39:57 Quit lovasoa (Read error: Connection reset by peer)
23:39:59 Join lovasoa3 [0] (~yaaic@
23:40:24 Quit lovasoa3 (Remote host closed the connection)
23:43:52 Quit lovasoa2 (Ping timeout: 245 seconds)
23:54:08 Quit domonoky (Read error: Connection reset by peer)

Previous day | Next day