<TheSeven>
hm, android thinks that sunxi-ir is a hardware keyboard and refuses to show a soft keyboard...
<Sonic1>
hah
tzafrir has joined #linux-sunxi
<nfet_>
Hi, where the best source for finding uboot image or source for A31?
<nfet_>
I'm bringing up bare metal A31
<ssvb>
libv: replied to that thread in the ml
<ssvb>
I sincerely hope that aseigo was not attempting to just troll us
<ssvb>
even though "i’m sure someone is running to send an email containing the words “volunteer effort”, “community” and what-not at this point. ;)" suggests otherwise
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
aesok has quit [Remote host closed the connection]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
prasannapete has joined #linux-sunxi
wingrime has joined #linux-sunxi
parabyte has joined #linux-sunxi
keebler has quit [Ping timeout: 240 seconds]
<parabyte>
I have looked hard and cant seem to find the original zet6221 module
booss has quit [Quit: Leaving]
<ganbold_>
wens: I'm confused, does cubieboard2 has RMII/RGMII that can do gigabit?
<wens>
ganbold_: no it doesn't, the phy only supports MII
<ganbold_>
wens: that is what I thought so, but looking your submitted patches confused, now it is clear :)
wingrime has quit [Ping timeout: 246 seconds]
discopig has joined #linux-sunxi
<wens>
:)
<parabyte>
i been using wingrime's driver
<parabyte>
but cant get anything out of it
<parabyte>
just not happening
<parabyte>
compiled source fine
<parabyte>
still no play
<parabyte>
xorg picks it up as evdev device
<parabyte>
no play
<parabyte>
i download latest git version of evdev input driver
<parabyte>
it still no play
<parabyte>
I think i found what is up with xorg and evdev
parabyte has quit [Quit: Leaving]
Gerwin_J has quit [Quit: Gerwin_J]
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
KBme has quit [Ping timeout: 252 seconds]
Black_Horseman has quit [Remote host closed the connection]
<ssvb>
but at least crypto is much simpler than 2D "acceleration" (which is also a total disaster in every ARM based device out there), so crypto can be used as a training exercise before moving to more complex things
kivutar has quit [Ping timeout: 240 seconds]
<tomee^>
ha
<tomee^>
got that godforsaken bt chip on cubietruck to work
<mnemoc>
:o
<tomee^>
at least it pairs with my mobile
<mnemoc>
\o/
<tomee^>
haven't tested all those nifty profiles yet, though
<tomee^>
ssvb: I thought video decoding and acceleration was a disaster in the ARM devices
<tomee^>
ssvb: ...until I got my new laptop last week
<tomee^>
ssvb: now, how THEY managed to complicate things is just amazing.
<tomee^>
btw could someone with git experience tell me:
<tomee^>
say I have messed around in a cloned git tree (which I did)
<tomee^>
and in the meantime the upstream had perhaps changed
<tomee^>
is there a way to generate a diff of my changed files against the revision that I pulled and convert that into a 1) patch 2) pull request 3) fork ?
<ssvb>
tomee^: video decoding is an entirely different thing
<Nyuutwo>
tomee^: you just commit, and then git pull --rebase which will try apply your commit on top of branch
<tomee^>
Nyuutwo: I don't need no rights to the original repo in the first place?
<tomee^>
Nyuutwo: see, I was raised in the times of RCS and CVS, and to commit you needed permissions first. that's different in git?
<Nyuutwo>
tomee^: only rights to read
<Nyuutwo>
tomee^: in git you have a mirror of whole repo on your disk
<Nyuutwo>
and you can commit it (locally)
<tomee^>
ok, I see
<Nyuutwo>
and then you can push changes to server
<Nyuutwo>
or when you have your server
<Nyuutwo>
other people can pull your changes
<tomee^>
hm
<tomee^>
so I commit my changes
<tomee^>
which I guess registers the changes that I made to the files under GIT control
<montjoie[home]>
thanks ssvb for the link
<Nyuutwo>
tomee^: yep
<tomee^>
and that's still on my hdd
<Nyuutwo>
tomee^: git gui if you want to see what will be commited
<tomee^>
now how do I make a pull request/branch out of that?
<Nyuutwo>
tomee^: in git you have working copy, staging, and repo
<tomee^>
Nyuutwo: heh, I guess I'll just have to RTFM :)
<tomee^>
Nyuutwo: it is totally different from svn
<Nyuutwo>
tomee^:
<Nyuutwo>
I haven't yet made any pull requests
<tomee^>
ssvb: well, video decoding in hardware is somewhat related to 2d acceleration, no?
<Nyuutwo>
but in genereal you publish your repo (for example on github)
<Nyuutwo>
and then you just say to person that your changes are avalible <link>
<Nyuutwo>
in github you have neat tool for merge request but it is not accepted globally
<tomee^>
Nyuutwo: ok, so in principle, I commit first, and then fork, and then push/checkin to the forked repo?
<ssvb>
tomee^: video decoding is a clearly separated independent task with its own input and output, which processes data in bulk
<Nyuutwo>
tomee^: generally yes
<Nyuutwo>
but
<tomee^>
ssvb: ...but the result of the hardware-aided decoding often bypasses all other pipelines, no?
<Nyuutwo>
firstly, get git server (account on github)
<Nyuutwo>
create repo
<ssvb>
tomee^: 2d acceleration has to deal with the sequences of arbitrarily small operations (such as when handling text glyphs)
<Nyuutwo>
link it to your PC git remote add
<ssvb>
tomee^: and there is also a large variety of these 2d operations, which sometimes require software fallbacks for the things not handled by hardware
<Nyuutwo>
make changes & merge localy with upstream
<Nyuutwo>
inform upstream about your patch
<tomee^>
ssvb: yeah, you're right. when put that way, 2d accel is indeed entirely different from hardware-aided video processing.
<Nyuutwo>
ah, before informing push changes to your repo with git push <remotename> <branchname>
<tomee^>
Nyuutwo: the thing is, I *already* made the changes. without having an account or informing git that I'd like it to track my hackery.
<tomee^>
Nyuutwo: but I think I get the idea, generally, thanks for your help.
<Nyuutwo>
tomee^: it is not a problem
<Nyuutwo>
just get server
<Nyuutwo>
and git remote add, git push <remotename>
<mnemoc>
bfree: can you see with libv to get kernel packages in packages.linux-sunxi.org/debian/ ?
<tomee^>
Nyuutwo: anyway - I just wanted to create a "branch" of a kernel for cubietruck. 99,99% linux-sunxi, plus cherry-picked ssvb's (hi, ssvb! :)) patches for mali r3p2 + my wifi hackery. might be useful for some people I guess.
<tomee^>
Nyuutwo: ok, and after I push, git will somehow magically track that it's been pulled from X and then offer a way to do a compare of the two repos?
<tomee^>
ssvb: btw I don't know if you are aware of this all "optimus" creepiness that is going on with PCs right now
<mnemoc>
tomee^: it makes more sense to submit them for stage/sunxi-3.4 :)
<tomee^>
mnemoc: they won't be accepted.
<tomee^>
mnemoc: for now the wifi stuff is a backport from cubietech.
<Nyuutwo>
tomee^: you just push whole kernel repo to github
<tomee^>
mnemoc: so it essentialy is a step backwards. but works... for now...
<mnemoc>
ok
<Nyuutwo>
tomee^: but when you make branch, there are 2 commits with the same parent
<tomee^>
Nyuutwo: err? ;)
<Nyuutwo>
tomee^: commits in git are linked with parents
<tomee^>
Nyuutwo: you mean "null" commits?
<Nyuutwo>
tomee^: null?
<tomee^>
Nyuutwo: as in making a branch generates 2 commits - one in the parent saying "branched into X" and the other in the newly-created branch saying "branched from X" ?
<Nyuutwo>
nope
<tomee^>
ok
<tomee^>
I'm off to the bookstore tomorrow
<tomee^>
;)
<Nyuutwo>
git saves only pointers to head of branch
<Nyuutwo>
and head have its parent ...
<Nyuutwo>
only first commit has no parent
<Nyuutwo>
and when you make branch you just make new label
<Nyuutwo>
but when you commit on branch, then you have 2 commits with the same parent
<bfree>
mnemoc: what kernel packages?
<mnemoc>
the ones you make :)
<tomee^>
Nyuutwo: ok, now I think I understand. I mixed up branches with forks.
<bfree>
mnemoc: next ones I make will probably be 3.13ish ones ... I don't think libv will have any interest in those ;)
<Nyuutwo>
tomee^: forks on github are clones of repo in your account
<hno>
tomee^, what is a fork in your view?
<hno>
git don't have forks, only different heads..
<hno>
all the time-
<tomee^>
hno: I have no idea. I guess a fork in git makes it stop tracking changes against branches of parent, tracks only between master?
<hno>
tomee^, no.
<tomee^>
hno: so github technically tracks all diffs? like N^N complexity? wow.
<hno>
a fork has nothing to do with tools or tracking. It's political. When a project splits in different directions.
<hno>
what github calls forks is a clone, not a fork. It may or may not be the start of a fork.
<hno>
tomee^, it's git that tracks them. github just stores them and makes some nice graphs and statistics from them.
<mnemoc>
it's sometime weird that in github's web interface you can access hashes from any "fork" on any related repo