**** BEGIN LOGGING AT Mon Feb 15 02:59:59 2016 Feb 15 19:57:42 http://k12cs.org/review/ <= couple days left for current review period Feb 15 20:51:23 Hi. I'm interested in kernel stuff. I actually just got my first patch accepted (into gregkh's staging tree) yesterday. And it was a small cleanup patch, so I don't have much experience with kernel development yet. The project ideas page still looks a little sparse. Being that the organization application is due later this week I believe, should I expect to see some ideas start popping up? Or do the st Feb 15 20:51:29 udents generally come up with the ideas? Feb 15 20:53:41 I do have hardware experience (not with the Beaglebone though), so I'd be good with project ideas also involving hardware or not, in addition to software (though I guess kernel projects would also involve hardware of course). Feb 15 20:57:59 <_av500_> jbisch: gsoc is mostly about SW Feb 15 20:58:35 <_av500_> so even we are HW, the gsoc projects need to have a clear SW focus Feb 15 20:58:38 <_av500_> kernel work is great Feb 15 21:00:48 _av500_: Okay, thanks. Feb 15 21:02:20 <_av500_> jbisch: and yes, we do also accept student project ideas Feb 15 21:15:54 please suggests ideas Feb 15 21:16:12 it should be beneficial to the greater beagle community Feb 15 21:23:02 I'm looking at the 2015 ideas for things that haven't been done and saw the upstreaming kernel patches one. I figure it might be good practice to try doing a little. I see the old repo that is linked to kept the patches separate from the kernel and used a script to apply them, but the current repo appears to have the patches already applied to the kernel tree. So what would be the best way of determini Feb 15 21:23:09 ng commits that aren't in the mainline kernel yet? Would "git format-patch linus-tree..bb-tree" be the best option? Is this still something that is needed? Feb 15 21:23:36 And how is this organized? Feb 15 21:24:01 I mean is there someone to corrodinate with if I wanted to do some of that? Feb 15 21:29:31 jbisch, for upstreaming... the big ones are remoteproc_pruss (some of which is still in development at ti.com), overlays, we are down to about 40 patches that need to be pushed.. Feb 15 21:30:25 jbisch, this is the current in-flight patchset https://github.com/RobertCNelson/linux-dev/blob/master/patch.sh#L329-L796 Feb 15 21:31:07 jbisch, note i also track stuff to backport to lts-4.4 in that same section.. Feb 15 21:32:40 rcn-ee: Okay. So since the patches already exist, this involves making sure the patches are up to the standards of the mainline kernel and they are checkpatch clean? Feb 15 21:33:05 yeah, that's about 99% take care of. ;) Feb 15 21:33:53 a lot of what's left is more about the, politic's pro/con's etc... the really fun stuff about pushing upstream. ;) Feb 15 21:39:47 rcn-ee: I guess I'll try taking a look at that first group (The 7 Convert exiting EEPROM drivers to NVMEM). Should I run the patches by you before emailing the appropriate mailing list? Or just CC you on the mailing list post? Feb 15 21:41:02 jbisch, so anything with #[PATCH X/Y], was something i pulled from a mailing list.. With that specific one, there was one bug, so V2 should show up at some point .;) Feb 15 21:41:27 (i'm expecting that one to hit 4.6-rc0) Feb 15 21:44:18 So something like 0001 in bb_overlays would be better to look at now. Feb 15 21:45:28 yeah, panto-cloud's working on 0001-OF-DT-Overlay-configfs-interface-v6.patch Feb 15 21:45:58 the 3 eqep's might the easist to get upstream.. ;) Feb 15 21:46:52 pinmux-helper can be ignored, trying to make the "connector" patch take that over.. Feb 15 21:47:10 the adihdmi board isn't avaiable yet.. Feb 15 21:47:15 Okay. Then I'll look at the 3 eqeps. Feb 15 21:48:15 for the eqep's the original author should be cc'ed, i don't know if he cares about that patch anymore.. the author of 2/3 is using it in production.. Feb 15 21:48:56 i can image it needs to be re-written for 'iio' subsystem as it sorta fits that. Feb 15 21:51:11 Only 292 checkpatch errors in 0001 :o Feb 15 21:52:22 jbisch, yeah that one was bad... Then Dick (patch 0003) fixed most of those along with tweaks.. (yeah i should have cleaned up 0001 a few years back() Feb 15 21:55:22 rcn-ee: So once I'm done, do I sign off on the patch? I'm not sure what to do if I'm just upstreaming someone else's patch? Feb 15 21:58:59 make sure to leave his name in first, and reference him in the commit, then when you use git send-patch make sure to cc him... That way he can add his sign-off Feb 15 21:59:38 but i have a feeling upstream will want you to re-write it for the iio subsystem, as there are encoders there now.. (there wasn't way back in 3.8.x days when that was first written) Feb 15 21:59:57 https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/iio Feb 15 22:01:12 Alright. I'll see what they say with it not rewritten and then deal with rewriting if they want that. Feb 15 22:02:24 and you can squash my 0002 into 0001, that was just an upstream interface change.. Feb 15 22:49:04 ds2@acer0 /nas2/hyr/datasheet/TI/Processor $ ls Feb 15 22:49:55 blah Feb 15 22:50:04 * ds2 chants fork the kernel **** ENDING LOGGING AT Tue Feb 16 02:59:59 2016