[03:24] <slangasek> lfaraone: opting out of the HWE stack == use 12.04.1 (or 12.04.0) media for install
[03:27] <geofft> slangasek: these are nontechnical end users, they click the thing that says "Ubuntu", we don't get to tell them what to install until after they installed :-( 
[03:28] <geofft> we put up a warning on our home page, but people will still go and install Ubuntu and then say "oh hm I want AFS" later 
[03:28] <geofft> and they get the default obvious CD image from ubuntu.com 
[03:29] <geofft> ... actually _I_ can't figure out how to get 12.04.x for x<3 from http://www.ubuntu.com/download/desktop 
[03:30] <geofft> alternative downloads | past releases takes me to releases.ubuntu.com, and that just give sme 12.04.3 
[03:30] <infinity> geofft: Bottom of the page, "Take a look at a full list of our previous versions and alternative downloads".  And agreed, that's not even remotely intuitive.
[03:31] <geofft> infinity: I don't think it's there! 
[03:31] <infinity> geofft: It is, bottom of THAT page, "past releases..."
[03:32] <geofft> keep clicking. it's not there. 
[03:32] <infinity> Which... Then goes to releases.u.c, which remains unhelpful, you're right.
[03:32] <infinity> *sigh*
[03:32] <geofft> Oh! I go to "unsupported Ubuntu images". 
[03:32] <infinity> For the record, it's on old-releases.
[03:32] <geofft> That gets me 12.04.0 for ... powerpc, nevermind 
[03:33] <infinity> The link right under the one you clicked. :P
[03:33] <infinity> Anyhow, we need to make this experience far less crap.  I have a todo item for that.
[03:33] <geofft> Yeah, I am going to politely disagree with the claim that it's possible for an end user to figure this out. We're in the position of downgrading kernels post-install 
[03:34] <geofft> Or, really, just having people install OpenAFS from an alternative source, which is what we're doing now. (I think our installer shell script sets up the PPA) 
[03:34] <infinity> Well, it's *possible*, but I'll agree that it's not *plausible* that most will figure this out.
[03:35] <geofft> And yes, afaik (which is not very rigorously, to be fair), our users tend not to have hardware that requires the HWE stack 
[07:41]  * apw whines
[08:59] <xnox> slangasek: the fact that ubiquity installer doesn't show anywhere which version of ubuntu is installed, is not helpful either =)
[13:40] <pkern> Hey. https://help.ubuntu.com/community/Kernel/Compile#Modify_the_source_for_your_needs seems a bit outdated. If I have a linux-lts-raring apt-get source'd, how would I modify the kernel config? Somehow all files I find have a splitconfig.pl header.
[14:58] <apw> pkern, add the option you want to change to the apporpriate leaf file in the config, and then run 'fakeroot debian/rules updateconfigs'
[14:58] <apw> which will sort out making the configs consistant
[15:55] <ppisati> bug 1250495
[15:55] <ubot2> Launchpad bug 1250495 in flash-kernel (Ubuntu Trusty) "armhf: highbank: relocate initrd to make room for a bigger kernel" [Undecided,New] https://launchpad.net/bugs/1250495
[15:55] <ppisati> infinity: ^
[15:56] <ppisati> infinity: i just met the same problem with a T kernel on midway, any chance you can give it a look?
[15:57] <infinity> ppisati: Yeahp, remind me again in ~4h, I'm running out to an airport (picking someone else up for once, not flying anywhere myself).
[15:57] <infinity> Assuming I survive the trip there and back... The snow here is insane today.
[15:58] <ppisati> infinity: ack, i'll do
[15:58]  * ppisati is envious of your snow :P
[16:36] <jsalisbury> ##
[16:36] <jsalisbury> ## Kernel team meeting in 30 minutes
[16:36] <jsalisbury> ##
[16:51] <olli> hey, I am looking for a pointer how to catch a "swivel" event on a Dell XPS 12 (where the screen flips horizontally and turns it into a "tablet")
[16:51] <olli> I have tried to listen to /dev/input/event* and also looked in /proc/bus/input/devices for pointers
[16:52] <olli> is there another common place to look at?
[16:52] <sforshee> olli: I'd expect dell-wmi to be receiving the events. If it's an event dell-wmi understands it will probably show up on the input device it creates.
[16:52] <sforshee> olli: otherwise it would need to be taught about the event
[16:53] <sforshee> olli: but you might also try running acpi_listen and see if it spits out anything
[16:54] <olli> sforshee, ah, forgot to mention that acpi_listen didn't produce anything
[16:54] <olli> sforshee, is it a valid approach to just cat /dev/input/event* and see if something happens or would cat miss something?
[16:54] <olli> event* = a single event dev, e.g. event10
[16:55] <sforshee> olli: that might not work. X grabs the input devices so that nothing else sees the events.
[16:55] <sforshee> olli: what I usually do is install input-utils, swtich to a vt, use lsinput to identify the device I'm interested in, then use input-events to watch the events from that device
[16:56] <olli> sforshee, awesome, thanks, I will try that
[16:58] <sforshee> olli: the other thing I'd suggest is seeing if anything appears in dmesg when you rotate the screen
[16:59] <jsalisbury> ##
[16:59] <jsalisbury> ## Meeting starting now
[17:00] <jsalisbury> ##
[17:00] <sforshee> olli: one of the relevant messages is debug-level and won't appear in dmesg by default. Let me dig up the command that will enable it.
[17:01] <sforshee> olli: echo "module dell-wmi +p" | sudo tee /sys/kernel/debug/dynamic_debug/control
[17:01] <sforshee> that should do it
[17:12] <olli> sforshee, that doesn't seem to trigger any additional messages, and using input-events on the dell-wmi device on a vt doesn't show anything either
[17:13] <olli> is this the case you mentioned earlier where dell-wmi needs to be taught about it
[17:13] <sforshee> olli: no, if it received events that it didn't know what to do with you should have seen something in dmesg
[17:13] <olli> hm
[17:14] <sforshee> I'm not coming up with anywhere else to look off the top of my head
[17:14] <olli> sforshee, then there is still a big chance of pebkac on my side, will keep trying
[17:14] <olli> sforshee, thx for now, might bug you later
[17:15] <sforshee> olli: are you absolutely sure it's generating an event? I.e. is there some observable behavior in Windows which leads you to that conclusion?
[17:18] <olli> sforshee, http://www.youtube.com/watch?v=nU9Pa3nF6y4 at 0:50 you can see windows switching modes when swiveling, I of course nuked my win partition, but will install to verify
[18:23]  * rtg -> lunch
[20:54]  * rtg -> EOD
[21:18] <pkern> apw: So debian.master and it will do the appropriate change to debian.raring? Or are LTS backports even more special?
[21:46] <apw> pkern, lts-backports are even more special
[23:26] <pkern> apw: Ok, so I guess I'll try to modify debian.raring tomorrow. I need DYNAMIC_DEBUG for some e1000e suspend issue. |: