[09:44] <GeorgeJ> Hello folks!
[09:44] <GeorgeJ> IS there any ARM channel on freenode?
[09:44] <GeorgeJ> Specific to ARM, not neccesarely ubuntu-arm.
[09:52] <cooloney> GrueMaster and NCommander, i uploaded armadaxp kernel to PPA, building is ok.
[19:11] <NCommander> infinity: ogra_: anyone want to review the kernel before I upload?
[19:12] <infinity> NCommander: It'll land in NEW anyway, I can review it there. ;)
[19:13] <infinity> NCommander: Unless you suspect it's a mess and will need a few iterations.
[19:14] <NCommander> infinity: its pretty hidious, but I think I got everything correct.
[19:14] <infinity> Is there any reason that, despite using the omap4 packaging, it's only building for armhf?
[19:14] <GrueMaster> infinity: Define "mess".  It does need work, but the sooner we get it into main, the sooner we can spin images and hammer it with more tests (and file bugs against it).
[19:14] <infinity> I mean, I know we'll only build IMAGES for armhf, but there's no reason not to provide the kernel for armel too.
[19:15] <GrueMaster> Mainly for testing.  We have a very limited quantity of systems.  Beyond that, I don't see a reason for no armel rev.
[19:16] <infinity> NCommander: Anyhow, what I see for ~ppa4 looks fine on the surface, but I haven't done copyright audits and the like, which I'll do when it lands in NEW.  So, make sure debian/copyright is sane.
[19:16] <infinity> GrueMaster: Yeah, it just didn't make sense to me to intentionally fork that from the omap4 packaging.  Had it been cargo-culted wholesale, it would have armel too. ;)
[19:16] <infinity> But, not picky.
[19:17] <NCommander> infinity: well, I need to cut some crud, I don't thinkwe need linux-armadaxp-libc-dev
[19:17] <infinity> Most people should only ever want armhf on this thing anyway.
[19:17] <infinity> NCommander: I don't see that in the published packages...
[19:17] <NCommander> oh, it got stubbedout
[19:17]  * NCommander is going through one last time
[19:19] <infinity> NCommander: Anyhow.  Yeah.  Just double-check that licensing stuff is right, blah blah, any last-minute crap you care about, upload, and I'll put on a different hat and give it a proper review.
[19:19] <infinity> NCommander: And we did, ultimately, get agreement from all concerned parties to jam this in main, right?
[19:19] <NCommander> infinity: I used the existing copyright with a notice that its Linux for ArmadaXP, and that Marvell added that support
[19:19] <NCommander> infinity: yes, we did, but we need a proper MIR for it
[19:19] <infinity> We... Do?
[19:19] <NCommander> (per skaet)
[19:19] <infinity> It's a kernel.
[19:20] <infinity> I'll talk to her about that.
[19:20] <NCommander> infinity: publish it to universe for now. We can promote it later, but I need something sitting in archive
[19:20] <infinity> I'll talk in parallel with my review.
[19:20] <infinity> Of the tlak goes nowhere fun, I'll accept to universe.
[19:20] <NCommander> infinity: thanks
[19:20] <infinity> But I'd prefer to accept to main.
[19:20] <infinity> So, we'll see how that goes.
[19:21]  * NCommander would like someone else review this before kicking to the archive, but uploading 100 MiB to REVU on this linux will SUCK
[19:21] <NCommander> s/linux/link/g
[19:22] <infinity> And no orig either.  So, every iteration is a new land of suck.
[19:22] <NCommander> infinity: I built one
[19:22] <infinity> But, honestly, all of our !primary kernel packaging is a mess.
[19:22] <NCommander> or more specifically
[19:22] <infinity> apw and I have been talking about ways to make that less so.
[19:22] <NCommander> I grabbed the .orig that was used in oneiric
[19:22] <infinity> But for now, "Whatever, if it builds packages that work."
[19:23] <NCommander> but it should make future uploads be about 4 MiB (the Marvell crap lives in diff.gz :-()
[19:26] <NCommander> infinity: uploading now. Holding about 200 KiB (I found T-Mobile gave me better upstream than my home ISP or starbucks)
[19:28] <infinity> That's a bit sad.  But good advertising for tmo. :P
[19:30] <NCommander> 3 Mbit up/7 down according to speed test
[19:30] <NCommander> Home connection is 256kpbs up/25 down
[19:30] <NCommander> and t-mobile offers $30 prepaid SIMs with 5 GiB. When I hit the bandwidth cap, I run to the store, buy a new SIM, and run with it
[19:31]  * NCommander went through about 10 GiB of 3G data in a week when we were in Austin due to the pathetic hotel wireless
[19:38] <NCommander>   Uploading linux-armadaxp_3.0.0.orig.tar.gz: 19570k/94410k[Errno 104] Connection reset by peer
[19:38] <NCommander> ARGH$#!@#!#^$
[19:39] <infinity> NCommander: rsync --partial to chinstrap, and dput from there.
[19:39] <infinity> NCommander: Then you can at least resume.
[19:41] <NCommander> good idea
[19:43] <GrueMaster> NCommander: I already have the ppa4 kernel in my chinstrap directory, if you want to save time.
[19:43] <NCommander> GrueMaster: cooloney packaged it wrong and there is no orig.tar.gz so it needs a reupload
[19:44] <GrueMaster> bah.  Figures.
[19:44] <infinity> NCommander: Well, also...
[19:44] <infinity> NCommander: If that's a bit-for-bit copy of the orig from oneiric, just grab it from the librarian to chinstrap and rename it.
[19:44] <infinity> NCommander: Then upload diff and dsc, and dput.
[19:44] <infinity> diff and dsc and changes even, but you know what I mean.
[19:45] <NCommander> infinity: indeed
[19:45] <GrueMaster> infinity: I didn't push the source up.  Just the binary .deb
[19:46] <GrueMaster> Let me know if there is anything I can do to help.
[19:47]  * infinity grabs a snack while this all goes on.
[19:51] <NCommander> uploaded
[19:51]  * NCommander waits for a NEW email
[19:59]  * NCommander still waits for a NEW email
[20:03] <NCommander> infinity: [ubuntu/precise] linux-armadaxp 3.0.0-1500.1 (New)
[20:03] <NCommander> do your thing
[20:16] <infinity> NCommander: Thinging.
[20:23] <GrueMaster> ewww,
[23:52] <TheMuso> ogra_: That thread for the AC100 config patch didn't go anywhere, so if that alsa-lib config snippet works with the aC100 for the kernel you have in precise for that hardware, then I'll put it in for precise, and we can sort it out properly in the future, when are get into the device-tree ira.