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 2011-07-01

00:01:43 Quit dantje_ (Quit: Ex-Chat)
00:01:59 Join Thra11_ [0] (
00:03:01 Quit Thra11 (Ping timeout: 252 seconds)
00:06:13 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 3.6.18/20110614230723])
00:06:25 Quit mudd1 (Quit: Ex-Chat)
00:12:38 Quit krazykit (Ping timeout: 258 seconds)
00:18:11 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
00:21:31***Saving seen data "./dancer.seen"
00:21:51 Quit Farthen (Read error: Operation timed out)
00:22:06 Join Farthen [0] (
00:25:50 Quit ender` (Quit: /[,\(\s]?#1(s)(\s+-?\d+(?{}\.\d+)?\s+-?\d+(?{}\.\d+)?\s+-?\d+(?{}\.\d+)?\s+-?\d+(?{}\.\d+)?\s+-?\d+(?{}\.\d+)?\s+-?\d+(?{}\.\d+)?)(\s+-?\d+(?{}\.\d+)?\s+-?\d+(?{}\.\d+)?)+(\)?=|\s|$)/)
00:27:57 Quit alexbobp (Ping timeout: 252 seconds)
15:02:29***Started Dancer V4.16
15:02:29***Connected to on port 6667
15:02:29***Logfile for #rockbox started
15:02:29Mode"logbot :+i" by logbot
15:02:33***Server message 501: 'logbot :Unknown MODE flag'
15:02:33 Join logbot [0] (
15:02:33 Join Thra11 [0] (
15:02:33 Join wodz [0] (
15:02:33 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
15:02:33 Join dan_a [0] (d917a37d@gateway/web/freenode/ip.
15:02:33 Join eGen_ [0] (
15:02:33 Join dfkt [0] (dfkt@unaffiliated/dfkt)
15:02:33 Join MethoS- [0] (~clemens@
15:02:33 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
15:02:33 Join n1s [0] (~quassel@rockbox/developer/n1s)
15:02:33 Join swilde [0] (
15:02:33 Join einhirn [0] (
15:02:33 Join quiccker [0] (~quiccker@
15:02:33 Join TheLemonMan [0] (
15:02:33 Join dionoea [0] (
15:02:33 Join utanapischti [0] (
15:02:33 Join petur [0] (~petur@rockbox/developer/petur)
15:02:33 Join ender` [0] (
15:02:33 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
15:02:33 Join liar [0] (
15:02:33 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
15:02:33 Join sideral [0] (~sideral@rockbox/developer/sideral)
15:02:33 Join ReimuHakurei [0] (~reimu@
15:02:33 Join FoolOnHill [0] (
15:02:33 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
15:02:33 Join Topy44 [0] (
15:02:33 Join Rob2222 [0] (
15:02:33 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
15:02:33 Join pixelma [0] (quassel@rockbox/staff/pixelma)
15:02:33 Join amiconn [0] (quassel@rockbox/developer/amiconn)
15:02:33 Join bthomson [0] (
15:02:33 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
15:02:33 Join bzed [0] (
15:02:33 Join alexbobp [0] (~alex@
15:02:33 Join balintx [0] (
15:02:33 Join FOAD [0] (~dok@
15:02:33 Join krazykit [0] (~krazykit@
15:02:33 Join Farthen [0] (
15:02:33 Join freqmod [0] (~fredrik@2001:700:300:1430:226:18ff:fe81:1e24)
15:02:33 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
15:02:33 Join boghog [0] (~aphax@2001:980:34c7:0:1e6f:65ff:fe86:1e03)
15:02:33 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
15:02:33 Join knittl [0] (~knittl@unaffiliated/knittl)
15:02:33 Join simabeis [0] (
15:02:33 Join avacore^ [0] (
15:02:33 Join zchs [0] (
15:02:33 Join AlexP [0] (~alex@rockbox/staff/AlexP)
15:02:33 Join n17ikh [0] (
15:02:33 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:02:33 Join belak [0] (belak@subtle/user/belak)
15:02:33 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
15:02:33 Join Pikel [0] (
15:02:33 Join sega01 [0] (
15:02:33 Join Xerion [0] (
15:02:33 Join factor [0] (~factor@
15:02:33 Join mikroflops [0] (
15:02:33 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
15:02:33 Join ack [0] (
15:02:33 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
15:02:33 Join CIA-14 [0] (
15:02:33 Join Zarggg [0] (
15:02:33 Join bieber [0] (
15:02:33 Join [fred] [0] (
15:02:33 Join jordan` [0] (gromit@2a01:e34:eebf:c890:21a:4dff:fe63:6966)
15:02:33 Join ender| [0] (
15:02:33 Join Xerrao [0] (~XRO@
15:02:33 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
15:02:33 Join guymann [0] (
15:02:33 Join simonlnu [0] (gfpg5UNKrx@unaffiliated/simonrvn)
15:02:33 Join JesusChrysler [0] (
15:02:33 Join Torne [0] (~torne@rockbox/developer/Torne)
15:02:33 Join Guinness [0] (
15:02:33 Join soap [0] (~soap@rockbox/staff/soap)
15:02:33 Join scorche [0] (~scorche@rockbox/administrator/scorche)
15:02:33 Join bug2000 [0] (~bug@unaffiliated/bug2000)
15:02:33 Join kkit|sh [0] (
15:02:33 Join niekie [0] (~niek@CAcert/Assurer/niekie)
15:02:33 Join Vimk [0] (
15:02:33 Join quem [0] (
15:02:33 Join tmzt [0] (
15:02:33 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
15:02:33 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
15:02:33 Join mystica555_ [0] (
15:02:33 Join parafin [0] (
15:02:33 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
15:02:33 Join mystica555 [0] (
15:02:33 Join Elfish [0] (amba@2a01:4f8:100:90a1:abc:abc:abc:abc)
15:02:33 Join Hadaka [0] (
15:02:33 Join gammy [0] (
15:02:33 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
15:02:33 Join jfc [0] (
15:02:33 Join zu [0] (
15:02:33 Join ruskie [0] (ruskie@sourcemage/mage/ruskie)
15:02:33 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
15:02:33 Join ranmachan [0] (ranma@2a01:4f8:130:9321::2)
15:02:33 Join pjm0616 [0] (~user@
15:02:33 Join TBCOOL [0] (
15:02:33 Join Slasheri [0] (miipekk@rockbox/developer/Slasheri)
15:02:33 Join Zambezi [0] (Zulu@
15:02:33 Join B4gder [241] (~daniel@rockbox/developer/bagder)
15:02:33 Join jae [0] (
15:02:33 Join rudi_s [0] (
15:02:33 Join aevin [0] (eivindsy@unaffiliated/aevin)
15:02:33 Join Espreon [0] (~espreon@wesnoth/developer/espreon)
15:02:33 Join rasher [0] (~rasher@rockbox/developer/rasher)
15:02:33 Join neferty [0] (~andor@
15:02:33 Join yosafbridge [0] (
15:02:33 Join Barahir [0] (
15:02:33 Join maraz [0] (
15:02:33 Join preglow [0] (thomj@rockbox/developer/preglow)
15:02:33 Join linuxguy3 [0] (
15:02:33 Join Utchy [0] (
15:02:33 Join @ChanServ [0] (ChanServ@services.)
15:02:33 Join desowin [0] (~desowin@atheme/member/desowin)
15:02:33 Join feisar-_ [0] (
15:02:33 Join iq [0] (~iq@unaffiliated/iq)
15:02:33 Join Staphylo [0] (
15:02:33 Join scorche|sh [0] (~scorche@rockbox/administrator/scorche)
15:02:33 Join ThomasAH [0] (
15:02:33 Join ved [0] (
15:02:33 Join Galois [0] (
15:02:33 Join merbanan [0] (
15:03:03 Part quem
15:04:35pamaurythe fuze+ uses SW i2c with HW i2c pins
15:05:20pamaury(if it helps you)
15:06:25B4gdersite is up
15:14:06 Quit krazykit (Ping timeout: 246 seconds)
15:28:58 Join fdinel [0] (
15:38:22CIA-14New commit by miipekk (r30108): Fixed incorrect dircache reallocation during tagcache commit.
15:41:46 Join patheticbliss [0] (
15:43:41 Quit wodz (Ping timeout: 240 seconds)
15:47:01 Join webguest608 [0] (
15:47:55 Quit webguest608 (Client Quit)
15:48:29CIA-14New commit by miipekk (r30109): Allocate indices directly in ramcache header. Add version code to ...
15:49:05pamaurythe website doesn't display commits
15:49:38Mode"#rockbox +o B4gder" by ChanServ (ChanServ@services.)
15:49:43Topic"3.9 released! | Please read before speaking: | Please direct offtopic/social chat to #rockbox-community | This channel is logged at | An upgrade caused issues with the wiki. We are on the case" by B4gder (~daniel@rockbox/developer/bagder)
15:49:47Mode"#rockbox -o B4gder" by ChanServ (ChanServ@services.)
16:01:44 Quit bthomson (Quit: WeeChat 0.3.3-dev)
16:05:27 Join bthomson [0] (
16:07:54 Quit petur (Quit: Ex-Chat)
16:15:02 Quit patheticbliss (Ping timeout: 276 seconds)
16:17:04 Quit bthomson (Quit: WeeChat 0.3.3-dev)
16:21:16 Join patheticbliss [0] (
16:21:42 Join bthomson [0] (
16:22:15sideraltest builds also don't show up
16:28:47CIA-14r30109 build result: All green
16:30:36Tornei assume songdbj in svn is not actually relevant or interesting to anyone any more, since it's not been changed since 2005? I was going to drop the branch for it (the code will still be present in trunk's history, since it was never changed after being split out of trunk)
16:32:24 Quit bthomson (Quit: WeeChat 0.3.3-dev)
16:33:10 Join asper [0] (
16:33:15asperhi guys
16:33:41asperI can send an Archos AV500 do some devs if someone is able to understand it and make a port for av500 !
16:33:47asperis anyone interested ?
16:34:08 Quit sideral (Disconnected by services)
16:34:08 Join sideral1 [0] (~sideral@rockbox/developer/sideral)
16:35:04 Join bthomson [0] (
16:35:10sideral1Torne: sounds good to me
16:35:19 Quit sideral1 (Remote host closed the connection)
16:36:47 Join sideral [0] (~sideral@rockbox/developer/sideral)
16:36:48sideralon the other hand, having an archival branch for it also wouldn't hurt
16:37:29gevaertsThe code there was last touched in 2005
16:37:36TorneWell, if we do want to keep a ref pointing at it, it'd probably be better to point it at the revision of trunk before it was deleted from trunk
16:37:40Tornerather than at the split-out branch
16:37:56Tornebecause the split-out branch includes duplicated commits due to being a subtree
16:39:20gevaertsTorne: I'd say drop it
16:39:46Tornei'm also dropping the branches for christi, dast, rockbox, zagor
16:39:49gevaertsIt seems to be a java tool to interact with our *previous* database
16:39:53Tornesince they contain no actual code that's different to what's in trunk
16:39:56sideralI'm not sure I fully understand this. What extra commits does it have if it was never changed after the split?
16:40:06Tornesideral: The things *before* the split have to be duplicated
16:40:15Tornebecause their effective root directory is trunk/songdbj
16:40:17Torneinstead of trunk
16:40:29Tornebecause in git branches/tags can't be subtrees of their parents
16:40:30sideralUgh. OK, drop it :)
16:41:13Torneanyway. the old cvs branches/tags for specific people are all empty of actual changes so i'm killing them and their associated tags
16:41:27Tornethemes, translate and www will all go into seperate repositories since they share no history with trunk
16:43:05Torneso that leaves bootloader_ams_pp, bootloader_ipod, gigabeat-s, mcdtexp, skinenginebreak and the v3_* branches
16:43:18Tornei assume everything in gigabeat-s, mcdtexp, and skinenginebreak is actually merged
16:43:31Torneand thus they don't really want to be active branch heads?
16:45:00aspersorry to be boring but... is there a reason why no-one port rockbox to archos AVXXX devices ?
16:45:19Torneasper: ports are done because people with the skills to do a port are interested
16:45:34 Quit sideral (Remote host closed the connection)
16:45:43Tornecontributing a device is rarely actually much help since generally if people are sufficiently interested to port to a device they either already have one or are willing to buy one
16:45:46Tornethough thanks for the offer :)
16:46:05asperok, sorry for my post :P
16:46:13Torneit's okay, we don't mind you offering :)
16:46:30Tornejust letting you know that it's not very likely someone is interested. they might still be :)
16:46:36asperI tested rockbox with archos recorder, sansa e200 aandroid and fuze
16:46:41asperit is a reallty GREAT software
16:47:06asperoh and also old archos jukeboxe player
16:47:28asperit seems that texas intruments chips are a pain in the ass for developers
16:47:38gevaertsasper: I'd say offers of hardware might be better on the -dev mailing list.
16:47:41asperthis may be is the reason why no one port it to av devices
16:47:43 Join sideral [0] (~sideral@rockbox/developer/sideral)
16:47:46pamauryasper: it also highly depends on the hardware
16:47:51Torneok, sod it, i'll write a proposal for what to do with our (rather large) list of branches and tags and post to the list
16:47:59Torneand if nobody objects i am right. ;)
16:48:29Torneactually one more questoin
16:48:55Tornegit generally doesn't distinguish branches and tags, you can use either in most places (and tabcompletion generally matches both)
16:49:12Tornethus there is potential confusion if we have a branch called v3.8 and also a tag v3.8
16:49:23Tornepossibly the branches should be called something different
16:49:30Tornesuggestions? :)
16:49:47TorneIt's nicer to keep the tags named that way
16:49:49Tornei think.
16:49:55Tornebecause git describe describes versions with respect to tags
16:49:58pamauryasper: by curiosity, what is the hardware of this archos ?
16:50:06pamauryyou are talking about a TI chip
16:50:08gevaertshm, ok
16:50:10Tornee.g. git describe <sha of v3.8> will return "v3.8"
16:51:00*gevaerts tries thinking about this
16:51:09gevaertsI think AlexP should decide
16:51:48dan_aTorne: Presumably a branch called "v3.8" would be working towards a 3.8 release? So call the branch "v3.8-rc" and the tag "v3.8"?
16:52:03Tornedan_a: heh.
16:52:10TorneThat's not a terrible idea, *but*..
16:52:12gevaertsdan_a: that would be misleading I think
16:52:22gevaertsThe branch also goes to 3.8.1 later on
16:52:38Torneone other thing I was intending to maybe do, was tag the trunk revision immediately after branching for 3.8 as v3.9pre or similar
16:52:59Tornesuch that describe will describe trunk versions as being v3.9pre-44-<prefix of sha>
16:53:01gevaertsv3.8-rb (release branch)?
16:53:04Torneif there are, say, 44 commits
16:53:14Tornewhich sort of preserves a sequential version number in a reasonably sane way :)
16:53:42gevaertsThat could be a good idea
16:53:51TorneYes, I think so
16:54:02TorneIt will include enough characters of the sha that it is guaranteed to be unique also
16:54:07gevaertsThat would actually fix the confusion of people thinking current svn is still some sort of 3.x
16:54:07Torneso you can ignore the number if you want
16:54:26asperthis is the page i compiled for archopen years ago:
16:54:31asperall the hardware documented
16:55:27asperthat seems to be a sort of prototype
16:55:33asperby TI
16:55:40Tornegevaerts: i think it's useful. we sort of discussed it at devcon and people were generally not against it
16:55:49 Quit sideral (Remote host closed the connection)
16:55:57Torneit's infinitely more useful to nondevelopers than a plain SHA
16:56:04*gevaerts nods
16:56:16Torneand it's less of a hack than trying to cobble together a sequential number ourselves by reading git log or whatever.
16:56:46 Join sideral [0] (~sideral@rockbox/developer/sideral)
16:56:49gevaertsOne thing we lose with git is the easy "modified" marker
16:56:50TheLemonManhow should i go to create a new rockbox target ?
16:56:55Tornegevaerts: describe can also do that
16:57:02Tornegit describe −−dirty
16:57:09gevaertsTorne: yes, but compared to what?
16:57:23TorneWell, there is that.
16:57:24asperno datasheet in 2008
16:57:26gevaertsThis "distributed" thing gets in the way :)
16:57:28TorneDirty is compared to HEAD
16:57:29aspernow I don't know
16:57:48TorneThere is an alternative way to construct git "versions" which identifies branches
16:57:53Tornebut it makes uglier looking output
16:58:01*Torne finds an example
16:58:17gevaertsasper: I believe we do have some TMS320DM320 code
16:58:56pamauryTheLemonMan: NewPort on the wiki ?
16:59:10TheLemonManthanks :)
17:00:59Torneok, git name-rev is useless because it goes backwards ;)
17:02:30***Saving seen data "./dancer.seen"
17:04:40Torneso hm this is actually a little hard ;)
17:04:56n1syeah, iirc the half working creative devices use that texas thing
17:05:51 Quit sideral (Disconnected by services)
17:05:52 Join sideral1 [0] (~sideral@rockbox/developer/sideral)
17:06:04 Quit sideral1 (Remote host closed the connection)
17:08:28 Join sideral [0] (~sideral@rockbox/developer/sideral)
17:09:39 Quit Thra11 (Quit: kthxbai)
17:14:59 Nick balintx is now known as nemvagyok (
17:16:19 Quit sideral (Remote host closed the connection)
17:16:58 Join sideral [0] (~sideral@rockbox/developer/sideral)
17:17:36Tornegevaerts: okay actually this is significantly less tractable than I thought :)
17:18:23 Nick nemvagyok is now known as balintx (
17:19:06Tornegevaerts: symmetry of distributed development kinda screws you.
17:24:58gevaertsTorne: maybe we can plan ahead and say how many revisions are allowed between releases? :)
17:25:45 Quit patheticbliss (Ping timeout: 260 seconds)
17:25:54Torneso yeah. describe with suitable tags works nicely for describing revisions on our trunk
17:26:04Tornebut it does not distinguish branches
17:26:34 Quit sideral (Remote host closed the connection)
17:26:47Torneand even if you explicitly check what branch you are on that doesn't stop it from describing something as v3.9-44 if it's a commit directly onto someone's local master that's the child of the (common) v3.9-43
17:27:21gevaertsWe'll have to learn to look up hashes I guess
17:27:24 Join sideral [0] (~sideral@rockbox/developer/sideral)
17:27:37Torneit's semi-tempting to suggest taht we make spit out only a SHA by default unless given some explicit arg telling it it's doing an "official" build
17:27:43Tornenot sure :)
17:29:14gevaertsWorth thinking about when we do the build script overhaul
17:29:33TorneThe downside there is that people doing their own builds will get nonsequential numbers
17:29:39Torneeven if they've changed nothing
17:29:52Torneunless they also do the same thing in which case you are back at not knowing if it's modified or not
17:30:23gevaertsDepends on how it works I guess
17:30:40Torneexpecting people to tell the difference between v3.9pre-44-g337e922 and v3.9pre-44-gf27028d is maybe tricky
17:30:46 Join patheticbliss [0] (
17:30:52Torneand even if you spot the difference you would have to go look up which one was "right" :)
17:31:24TheLemonManpamaury, how is the gpio direction set ?
17:31:50pamaurythere is no "direction", there is "output enable"
17:32:14gevaertsIf we go that way, we might want to consider something like v3.9pre-44-${SOMEIDENTIFIER}-g337e922, so it's not *all* just generated, and we set SOMEIDENTIFIER to "official" (or preferably something shorter...). That would solve the accidental confusion
17:32:27pamauryyou can read any pin whatever the function using DIN, but if you want it out, you must set it to GPIO and enable output to drive it
17:32:30 Quit ender` (Quit: The sardines were packed as tight as the coach section of a 747.)
17:32:36Tornegevaerts: That's probably better
17:32:48pamaury(and then enable/disable pull-up)
17:33:34gevaertsTorne: it would be good if the version string needs less than three minutes to scroll over a Player screen though :)
17:33:34dan_aTorne: Why not make ${SOMEIDENTIFIER} be the branch name?
17:34:01Tornedan_a: because in the "easy" case that may be trunk/master anyway
17:34:10Tornei certainly do not bother making a branch for every change I make :)
17:34:30Torneit needs to be something not automatic
17:34:35TheLemonManhuh in the i2c code there's no output enable access
17:34:51pamauryit's it bit-banging ?
17:34:54pamaury*is it
17:35:02Tornejust so that "builds from the build system" cannot be confused with similar looking builds from elsewhere which may have different sources
17:35:33Torneanyway, er
17:35:47Tornethe one I want to ponder really for now is what we call the branches we prepare releases on :)
17:36:09pamauryTheLemonMan: perhaps output is enabled by default, see the datasheet
17:36:13 Join ender` [0] (
17:36:23dan_aTorne: $VERSION-releaseprep?
17:36:31Torneother than "not vx.y-pre because we probably want to use that for describe output"
17:37:02Tornedan_a: possibly, but brevity is a virtue when some devices have small screens :)
17:37:38*dan_a must put Rockbox on his Sansa Clip at some point
17:37:48pamauryTheLemonMan: are you sure ? I say this because if you implement i2c by bit-banging, the difference between driven and not driven pin is very important
17:38:01gevaertsTorne: maybe use "v3.8" for the branch, and "v3.8-final", "v3.8.1-final" for the release tags?
17:38:23gevaertsThe latter can be long, since we override the version string for those anyway
17:38:36TorneAh, true
17:38:48pamauryperhaps the ROM enable the output for some reason
17:39:06 Quit sideral (Quit: Leaving.)
17:40:58Torneargh, git−−
17:41:54TheLemonMani fear the output is set somewhere in the first lines of start
17:42:28pamauryis that frightening ?:)
17:43:37TheLemonManhehe, tons of ldr/str in that part, getting confused is really easy
17:58:51 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:02:34 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
18:03:07 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
18:03:13 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
18:03:42 Quit n1s (Remote host closed the connection)
18:16:17kugelpHmm its too quiet to test my change to yaaic :-(
18:16:21 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
18:16:25 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
18:17:10 Join Horscht [0] (~Horscht@xbmc/user/horscht)
18:25:19 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
18:25:56mystica555kugel: heh
18:26:19 Join sideral [0] (~sideral@rockbox/developer/sideral)
18:26:46gevaertsTorne: have you looked at the bootloader_ipod history to see who to ask for more info about it? :)
18:38:09AlexPTorne: I think the v3.9-prexx is a fine idea - it is what e.g. quassel does and I also found it simple and useful
18:39:10 Quit dan_a (Ping timeout: 252 seconds)
18:39:32AlexPas for branches, maybe branch-3.8 or 3.8-branch or something like that? A bit ugly, but...
18:41:07gevaertsAlexP: I actually think I prefer the "v3.8" name for the branch and then "v3.8-final" or "v3.8-release" for the tags
18:41:39AlexPgevaerts: I agree, but I thought there was some reason I didn't understad for not doing that?
18:43:00 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
18:43:04gevaertsAlexP: I thought that this was the latest thinking, sweeping away all possible earlier objections
18:43:24AlexPok, I just briefly skimmed the log
18:43:59AlexPIf there are no technical reasons then I'd prefer v3.8 branches and 3.8-release tags or something similar
18:46:10AlexPah, I just read Torne's email
18:46:14AlexPSeems sensible
18:46:43pamauryTheLemonMan: did you keep disassembling the fuze+ ROM ?
18:49:55TheLemonMannop, i moved onto the bootloader quite immediately, ill care about the usb part later
18:50:16pamauryI'm reading the sb handling part
18:50:16 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
18:50:35 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
18:50:50pamaurythat's really painful to read :) And seem rather inefficient
18:51:55TheLemonManas 99% of their code heh
18:52:51amiconnTorne: We also need to keep the identifier reasonably short, otherwise it won't be fully visible on targets with small displays
18:53:18gevaertsTorne: can you move things between refs/heads and refs/oldheads later on without breaking things for other people?
18:53:25amiconnIt will scroll iirc, but deciphering something that's not just words while it's scrolling by isn't easy
18:53:54pamauryTheLemonMan: I mean, it goes 1) read 16-byte 2) interpret 1) 2) ... at least for the header part
18:53:58pamaurywe'll see for the rest :)
18:54:49amiconnOn archos bitmap there are 18 chars per line (rockbox default font), on the Clips, M3, and the iriver and iaudio remotes there are 21 chars. Player has just 11...
18:56:30gevaerts"v3.9pre-44-g337e922-official" would be too long to be really practical, on all of those...
18:56:55 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
18:57:11gevaertsThe "g337e922" is the most important part, so I think we can't really afford cutting that
18:57:52amiconnOur current scheme also includes the build date
18:58:16gevaertshm, right
18:59:01amiconnI'm not sure whether it's important with git - currently it is useful if you test something locally, as your base revision (from svn) won't change
19:01:00amiconnCurrent one uses 13 chars in total for an svn build including the 'M'
19:02:17 Join [Saint] [0] (~st.lasciv@
19:02:32***Saving seen data "./dancer.seen"
19:02:40 Quit bluebrother (Disconnected by services)
19:02:41 Quit [Saint] (Client Quit)
19:02:41 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
19:03:14 Quit TheLemonMan (Quit: Ex-Chat)
19:05:42amiconnB4gder/ Zagor: What's up with the build server?
19:06:07B4gderzagor isn't here, I have no idea
19:06:10B4gderwhat's wrong?
19:06:51amiconnThe table shows it took 48 minutes for building 30108, almost all with empty logs
19:06:57amiconn30109 wasn't built at all
19:07:30*amiconn also still misses the build system "kick" interface for committers
19:09:17amiconnHmm, only 2 clients??
19:09:30amiconnOne did 8 builds, the other did none?
19:11:50 Join adisbladis [0] (~user@2001:67c:2a0:9:216:3eff:fe6b:13eb)
19:13:23 Quit jordan` (Quit: Coyote finally caught me)
19:14:26Tornegevaerts: yes, you can move ref heads around
19:15:11gevaertsTorne: ok, in that case I'd move non-current release branches to oldheads as well
19:15:50Tornegevaerts: yeah, probably i guess.
19:16:30 Quit swilde (Remote host closed the connection)
19:16:54 Join fyrestorm [0] (
19:17:33Tornegevaerts: the way cloning works is you get everything under refs/heads, and you get everything under refs/tags *if* it refers to an object you actually downloaded
19:17:53Torneit doesn't download more history just because a tag points to it
19:18:00Torne(unless you tell it to)
19:18:18Torneyou can write your own refspec in your local config, which can map anything to anything
19:18:22gevaertsAnd refs isn't actually part of the hashed content then I guess
19:18:32TorneNo, refs are just.. symlinks, basically
19:18:48TorneThe actual tag objects are real objects
19:19:00Torne(so that they can carry descriptions/signatures/etc)
19:19:14Tornethe ref for a tag just points at the tag object, which then in turn points to the commit
19:19:41AlexPamiconn: I'm not sure date is so important here - it is useful currently as it is difficult to know immediately when an svn revision was, but with e.g. 3.9pre44 it is much more obvious.
19:37:17amiconnThe build date can wildly differ from the date a revision has been commited
19:38:04amiconnForinstance when you're bisecting
19:38:33AlexPyes, but does that matter?
19:39:01AlexPI guess what I'm asking is, is the build date useful for anything other than knowing when a current buuld that someone was using was current?
19:55:57 Quit Horscht (Read error: Connection reset by peer)
19:56:06 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:59:32 Join mudd1 [0] (
20:01:39 Quit asper (Quit: CGI:IRC)
20:07:09 Join Guinness` [0] (
20:09:02 Join stoffel [0] (
20:10:23 Quit Guinness (Ping timeout: 255 seconds)
20:10:36 Quit Guinness` (Read error: Connection reset by peer)
20:21:05 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
20:37:18 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
20:48:37 Join esperegu [0] (~quassel@
20:54:40 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
20:55:03 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
20:57:18 Join dan_a [0] (d917ad9c@gateway/web/freenode/ip.
20:58:11 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
21:02:35***Saving seen data "./dancer.seen"
21:02:58 Join [Saint] [0] (~st.lasciv@
21:18:22 Quit ReimuHakurei (Quit: If I use this, I will disappear, and Shana-tan will remain...)
21:19:47 Join ReimuHakurei [0] (~reimu@
21:23:14 Join Guinness [0] (
21:24:03 Quit MethoS- (Remote host closed the connection)
21:24:04 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
21:27:41 Join keyb_gr [0] (
21:40:07pamauryomg, I'm a fucking genius
21:46:22kugelppamaury: ...are you?
21:46:31pamauryyes definitely :)
21:46:33kugelpwhat did you invent? :)
21:47:09pamauryI just discovered something useful out of the freescale ROM disassembly
21:47:38pamaurythat undoubtly qualify as being a genius ;)
21:47:42 Join Strife89 [0] (
21:48:11pamauryand this also explains how the fuze+ boots, but that a minor point :D
21:51:53pamauryI did not understand how a boot section could pass control to another but the answer is simple: it gives a hidden argument to the entry point which is a pointer so that if you write the id of the next section, the ROM bootloader will branch to it
21:54:22pamauryI told you...
21:54:59 Quit [Saint] (Quit: Imagination is for turbo-nerds who can't handle how kick-butt reality is. I'm a kick-butt reality master! I would rather die, than be imaginative. I mean that.)
21:55:24pamaurySo actually, we will be be able to dual boot on the fuze+, the only *minor* point is that any OF update will destroy the bootloader...
21:56:37 Join [Saint] [0] (~st.lasciv@
21:56:40kugelpis that fixable?
21:57:10 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
21:57:10pamauryyes of course, by destroy I mean the rockbox bootloader will be gone, you will only be able to boot the OF and will need to reinstall it
21:57:51LloreanThat's the case on a lot of players though, so I don't think that's really a big deal.
21:58:08pamauryAnother problem is that if *by chance* the OF check the bootloader at each boot and overwrite it if it sees a difference, then you can't really call that dual-boot :)
21:58:49pamauryit would be a (rockbox-many/of-once)-boot
21:59:40LloreanThat's the case on the Gigabeat F (or was for quite a while) where it wasn't dual boot so much as "restore the OF bootloader if you want to get in the OF"
22:00:00pamauryAnother option is to use the i2c chip on the device but it's only 64kb large so fitting the bootloader in that would be a bit of pain I guess
22:01:47AlexPLlorean: I think it still is
22:01:55 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
22:01:56AlexPCouldn't swear to it though
22:03:31kugelp64k is plenty
22:03:38 Quit Strife89 (Ping timeout: 240 seconds)
22:04:02pamaurykugelp: with hardware init that becomes a lot less quickly
22:04:41kugelptwo stage loader perhaps?
22:05:53pamaurybut where do you store the second one ?
22:06:02kugelpI would think that a bl fits into 64k without bitmaps but I may be wrong
22:06:56pamauryI need to check this
22:06:57kugelppamaury: on disk?
22:07:43kugelpwell but then you can load rockbox.directly
22:09:39 Quit stoffel (Remote host closed the connection)
22:09:46pamauryyou still need to do the hardware init :) Nothing is impossible of course, but I just would like to make it as simple as possible
22:14:43 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
22:18:21 Quit [Saint] (Quit: Imagination is for turbo-nerds who can't handle how kick-butt reality is. I'm a kick-butt reality master! I would rather die, than be imaginative. I mean that.)
22:20:12 Join [Saint] [0] (~st.lasciv@
22:21:21 Join krazykit [0] (~krazykit@
22:33:59 Quit [Saint] (Quit: Imagination is for turbo-nerds who can't handle how kick-butt reality is. I'm a kick-butt reality master! I would rather die, than be imaginative. I mean that.)
22:35:28 Join [Saint] [0] (~st.lasciv@
22:38:06 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
22:38:26 Quit benedikt93 (Quit: Bye ;))
22:41:39 Join einhirn [0] (
22:42:10 Quit einhirn (Client Quit)
22:43:33amiconnpamaury: I don't know how much hardware init you have to do, but many rockbox bootloaders which have to do hardware init are smaller than 64K
22:45:49pamauryas I said, the problem is not much about the "if" but about the "how". Fitting it into 64kb would probably mean to rewrite the low level init stage which I extract from the OF
22:45:56pamaurywhich include the sdram init
22:46:00pamauryand which is a pain
22:46:49pamaurythat would probably save 20kb, plus the bitmap image and stripping useless thing you can probably fit this
22:47:26 Quit thomasjfox (Read error: Connection reset by peer)
22:50:22 Quit mc2739 (Ping timeout: 240 seconds)
22:52:27 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
22:55:47 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
22:55:59amiconnWhat bitmap? No rockbox bootloader contains a bitmap afaik (apart from the sysfont to be precise)
22:56:08 Join powell14ski_ [0] (
22:56:19pixelmaI believe some Sansas do
23:00:14 Quit tchan (Quit: WeeChat 0.3.5)
23:00:23pamauryyes it does
23:00:30pamaurybut you can disable it
23:00:47kugelamiconn: most do actually
23:02:38***Saving seen data "./dancer.seen"
23:03:26amiconnArchos bootbox doesn't, the iriver, iaudio, all ipod (<= 5.5G) bootloaders don't. c200v1 bootloader doesn't either
23:03:48kugelall sansas, abs i thought ipods too?
23:04:50kugelalso i seem to remember the change was applied to all bootloaders?
23:04:58amiconnThat's already 21 if I'm counting correctly
23:05:07kugelsome of probably were not released since then
23:07:02pixelmaI'm quite confident that it's not wanted or even possible on all due to size constraints
23:09:40 Quit [Saint] (Quit: Imagination is for turbo-nerds who can't handle how kick-butt reality is. I'm a kick-butt reality master! I would rather die, than be imaginative. I mean that.)
23:11:20 Join [Saint] [0] (~st.lasciv@
23:12:47sideralkugel: Have you taken note of jhMikeS' and my observation that the file browser shows trashed file names (one in each directory) when dircache is enabled?
23:13:15sideralThis is true even with Slasheri's latest commits
23:15:37 Quit [Saint] (Client Quit)
23:17:00 Join [Saint] [0] (~st.lasciv@
23:21:15 Quit [Saint] (Client Quit)
23:25:06 Join [Saint] [0] (~st.lasciv@
23:26:03 Quit esperegu (Read error: Connection reset by peer)
23:30:24 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
23:33:01 Quit [Saint] (Quit: Imagination is for turbo-nerds who can't handle how kick-butt reality is. I'm a kick-butt reality master! I would rather die, than be imaginative. I mean that.)
23:36:46 Join TheLemonMan [0] (
23:45:27 Join stripwax [0] (
23:57:41 Join tchan [0] (~tchan@lunar-linux/developer/tchan)

Previous day | Next day