=== mamarley_ is now known as mamarley
[11:31] <BenC> apw: FYI, the ethernet drivers for the freescale ppc hardware I use have finally sarted to hit mainline. I’d like to be able to reintroduce the kernel configs for them (two configs, one powerpc, the other powerpc64) for yakity.
[12:33] <apw> BenC, did we remove some flavours? 
[12:36] <BenC> I know e500 is gone, and that’s perfectly fine with me.
[12:36] <BenC> Speaking of flavours…
[12:37] <BenC> rtg: Does e500mc still get built?
[12:37] <BenC> Do you guys not use kernel.ubuntu.com git anymore?
[12:37] <apw> those names are not great, they are generic to all freescale things right?  a 32 and 64 bit variant thereof ?
[12:38] <rtg> BenC,  powerpc-smp powerpc64-smp powerpc-e500mc powerpc64-emb
[12:38] <BenC> Ok.
[12:39] <BenC> So the e500mc can be generalized a bit, but it doesn’t cover all 32-bit freescale (now nxp, btw)
[12:39] <BenC> e500 is soft-float, so needs a special kernel, but it’s also not used in any general production compete systems that I’m aware of.
[12:39] <rtg> BenC, we do still use kernel.ubuntu.com, but its not a mirror. the primary is on LP
[12:39] <apw> ugg, soft-float is even, die die die
[12:39] <BenC> powerpc64-emb variant is generalized enough for all book3e
[12:39] <rtg> s/not/now/
[12:40] <BenC> rtg: I didn’t notice that the git web listing is some 17 odd pages. I only looked at the first page
[12:40] <apw> BenC, yeah you should see all our trees on k.u.c but launchpad has tehm all now, and the cgit is quicker too
[12:41] <BenC> Anyway, e500mc can be renamed to powerpc-corenet, but I would need to verify the config has the correct things enabled to claim that
[12:41] <BenC> And I can verify that the fman drivers are enabled, since that’s improvement for the newer kernels
[12:42] <BenC> Took 5+ years, but they finally got their drivers upstream.
[12:43] <BenC> Any pull requests can be against: git://kernel.ubuntu.com/ubuntu/ubuntu-yakkety.git ?
[12:43]  * BenC wishes you guys would just use github
[12:44] <rtg> BenC, yeah, that is fine. I think the mirror runs every 30 minutes
[12:44] <apw> every 15m iirc, the master is
[12:44] <apw> https://launchpad.net/~ubuntu-kernel/ubuntu/+source/linux/+git/yakkety
[12:45] <xnox> BenC, you can create pull requests on launchpad =)
[12:46] <BenC> Shows how much I pay attention. I didn’t know lp had git support
[12:46] <xnox> BenC, that's been around for like a year, or maybe more, now.
[12:46] <rtg> BenC, hence the reason we don't use github.
=== Elimin8r is now known as Elimin8er
[12:49] <BenC> rtg: Are you ok with renaming powerpc-e500mc to powerpc-corenet in kernel and meta?
[12:49] <BenC> I’m fairly certain I am the only user
[12:49] <apw> what does "corenet" mean in this namespace ?
[12:50] <BenC> It cover a class or Freescale QorIQ SoCs that are powerpc based with frame manager components
[12:50] <rtg> BenC, its a bit late in the process, isn't it ?
[12:50] <BenC> It also covers a class of Freescale QorIQ SoCs that are arm based, but I don’t touch those...
[12:51] <BenC> rtg: I was just hoping to make it less specific for apw’s concern, but I don’t have a preference.
[12:52] <BenC> It will work for e500v2 and e500mc
[12:52] <apw> i was hoping to discover we would need just two variants going forward, always
[12:52] <apw> musch as we now do for powerpc (IBM style for want of a name)
[12:52] <BenC> Unfortunately, freescale SoCs are book3e and not book3s as IBM/Apple powerpc processors are.
[12:53] <BenC> So it will have to always be a seperate kernel build
[12:53] <apw> BenC, right, i mean 32/64 for each of "ibmish" and "freescaleish"
[12:54] <BenC> For 64-bit, it is and can remain that way (the e in book3e is embedded, hence the powerpc64-emb)
[12:55] <apw> and there is no hope of that for powerpc-emb ?
[12:56] <BenC> No. I’ve been down that road with benh and neither he, myself, nor anyone else, wants to perform a ppc32/mmu kenrel reorg to make that happen, and some even say that it can’t be done anyway (technically impossible).
[13:12] <apw> BenC, so no single -emb 32 bit flavour, sigh
[13:13] <BenC> Yeah, unfortunately, it’s not the type of thing that is resolved with device trees
=== ghostcube__ is now known as ghostcube
[13:41] <xnox> apw, i think i might be blind
[13:42] <xnox> I'm trying to find which git repository https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1572291 is committed to, if any
[13:42] <apw> xnox, ?
[13:42] <ubot5> Ubuntu bug 1572291 in linux (Ubuntu Xenial) "s390/pci: add extra padding to function measurement block" [Critical,Fix committed]
[13:42] <xnox> and i'm failing
[13:42] <xnox> i looked at https://code.launchpad.net/~ubuntu-kernel/+git
[13:42] <xnox> then found linux/+git/xenial and linux/+get/yakkety
[13:42] <xnox> yakkety doesn't have master-next branch
[13:42] <xnox> xenial master-next doesn't appear to have that commit
[13:43] <apw> rtg, ^ ?  he might of course be right in the middle of applying it
[13:43] <apw> xnox, oh but there it is on xenial master-next
[13:43] <xnox> wait maybe found it
[13:43] <xnox> https://git.launchpad.net/~ubuntu-kernel/ubuntu/+source/linux/+git/xenial/commit/?h=master-next&id=ab72705a2737c90c03d24b91afa10ef1b7403448
[13:43] <rtg> xnox, not doing anything with yakkety just yet
[13:44] <xnox> rtg, ok, cool.
[13:45] <xnox> i guess once there is yakkety kernel upload, the git forest of trees will look like typical bushes at that point.
[13:46] <apw> xnox, we work in xenail for a bit at this point and copy the kernel forard to yakkety, up till we have something which doesn't utterly stink
[13:46] <xnox> hehehe
[13:46] <xnox> =
[13:46] <xnox> =)
[13:47] <rtg> xnox, which could be 2-3 weeks yet
[14:26] <BenC> rtg: What’s the plan for yakkety’s kernel version? Is it moving past 4.4.x?
[14:27] <rtg> BenC, yup. I'm gonna get the latest one I can
[14:27] <apw> BenC, yep, it will move rather rapidly I am sure
=== Laney is now known as trmsu
[14:27] <rtg> maybe even an -rc if it is that close
[14:27] <BenC> Ok. The updated drivers I’m looking at are in 4.5+, so I’ll hold off until then to review the config options
[14:28] <BenC> 4.6 is at rc5 as of today. Seems plausible for inclusion.
=== willcooke is now known as wllck
[14:28] <apw> i'd be supprised if its not 4.7+
[14:30] <BenC> If you see CONFIG_FSL_FMAN come up in the config review, you’ll want to enable it for e500mc and ppc64-emb, but I’ll be sure to verify after you guys sync it up.
[14:30] <BenC> It’s possible you want it for some arm kernels as well, but I can’t comment directly.
[14:30] <rtg> BenC, you could have a look at git://git.launchpad.net/~ubuntu-kernel/ubuntu/+source/linux/+git/unstable which is what will become yakkety. I'm currently rebasing against 4.6-rc5
[14:33] <BenC> Thanks, I’ll take a peak
[15:06] <BenC> rtg: Things look good there
[15:06] <BenC> But this:
[15:06] <BenC> config/config.common.ubuntu:CONFIG_FSL_FMAN=m
[15:06] <BenC> Needs to be made =y
[15:06] <BenC> Wait, I’m wrong.
[15:06] <BenC> It can be loaded as a module, but it wont unload, so that’s fine as-is
=== trmsu is now known as Laney
=== clopez_ is now known as clopez
[16:39] <ben-8409> hello, i think a have a regression with latest kernel version in 16.04 (4.4.0-21) and nvidia drivers. I already filed a bug, but i wanted to check if there is anything more i can do? https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1574732
[16:39] <ubot5> Ubuntu bug 1574732 in xorg (Ubuntu) "Regression: Kernel Update in 16.04 from last days renders nvidia driver unusable" [Undecided,New]
[16:39] <ben-8409> using grub to boot into an older release (4.4.0-18-generic) fixes the issue for me.
[19:25] <kristian_on_linu> hi!
[19:25] <kristian_on_linu> how do I disable N support when configuring a kernel?
=== ghostcube_ is now known as ghostcube
[19:34] <kristian_on_linu> or, to be specific, 11n
[19:35] <kristian_on_linu> as mentioned here: http://askubuntu.com/questions/529347/how-do-i-keep-my-wifi-from-dropping-out