[07:20] <knightwise> morning
[07:21] <mapp> mornin knightwise
[07:25] <knightwise> hey mapp , how are you doing
[07:28] <mapp> not bad mate
[07:28] <mapp> you?
[07:28] <mapp> just got up for work?
[07:28] <knightwise> doin ok
[07:28] <knightwise> same here
[07:29] <knightwise> been up a couple of hours
[07:30] <knightwise> installed ubuntu on a mac mini in the house here
[07:30] <knightwise> going to be using it as a server for some vm"s and files
[07:35] <knightwise> looking for the best way to remote into it
[07:35] <knightwise> into the gui
[07:36] <mapp> what time did u get up if u been up a few hours!
[07:37] <knightwise> 5.30
[07:37] <knightwise> already did some digging in the yard before I start working from home
[07:37] <mapp> thats early:)
[07:38] <knightwise> used to it
[07:39] <knightwise> working from home this month because my co-workers are all on holiday
[07:39] <knightwise> i have the extra hours to do some work on the house and stuff
[07:41] <mapp> aha
[07:41] <knightwise> and run my own company :) (i'm an it consultant)
[07:42] <Gargoyle> knightwise: snap! (Also at home)
[07:42] <Gargoyle> taming the beast that is cloud-init. :)
[07:42] <knightwise> hey Gargoyle
[07:45] <knightwise> hmmm. . gonna try to get that mac mini up and running as a side project
[07:45] <Gargoyle> What you connecting from?
[07:46] <knightwise> Inside the network I would like to be able to connect using an RDP client
[07:47] <knightwise> under Gnome it used to be easy , under Unity , not so much.
[07:47] <Gargoyle> ah.
[07:47] <knightwise> i am willing to install lxde if required
[07:48] <Gargoyle> Used vnc in the past, because I mainly would be connecting from a Mac.
[07:48] <knightwise> Gargoyle: then what do you use on the linux side ?
[07:50] <Gargoyle> Most recently I've been using VM's that provide a vnc connection - but if I don't remember using anything elaborate before that. Just whatever google turned up!
[07:50] <mapp> was debating having another cig
[07:50] <mapp> but my throat hurts:(
[07:51] <Gargoyle> knightwise: vnc4server
[07:53] <knightwise> aha
[07:54] <knightwise> and that works on a headless machine (so no confirmation required and stuff like that ? )
[07:55] <Gargoyle> Or you could just use bash! :P
[07:56] <knightwise> i do use bash , everything I can do via the CLI , i do via the Cli
[07:56] <knightwise> but sometimes its handy to just have a quick glance at the desktop
[07:56] <Gargoyle> If it's a headless, why even bother with a desktop?
[07:58] <knightwise> I don't think you can run virtualbox without a gui
[07:58] <knightwise> at least not on the host machine
[07:58] <SuperMatt> tbh, I'd install a desktop so you can plug the thing in to a tv if you need to
[08:00] <knightwise> true SuperMatt , always nice to have it around even just if you are using it as a mediabox or something
[08:06] <SuperMatt> might not want to set lightdm/gdm to start automatically
[08:06] <knightwise> that I have (i installed it using the Mac64 desktop version)
[08:37] <bigcalm> Good morning peeps :)
[08:55] <knightwise> hey bigcalm
[08:59] <brobostigon> morning boys and girls.
[09:00] <knightwise> @hey brobostigon
[09:00] <brobostigon> morning knightwise
[09:02] <knightwise> hey brobostigon , how are you doing today
[09:05] <brobostigon> tired and boiled, and you knightwise ?
[09:07] <knightwise> working from home today
[09:08] <knightwise> still have 120 bricks , 500 kg of steel rods and about 1 cubic ton of cement to carry to the back yard
[09:08] <knightwise> thats something to do for this evening
[09:08] <brobostigon> what are you building?
[09:09] <knightwise> garden wall between me and my neighbour collapsed
[09:09] <knightwise> Time for a new one
[09:09] <knightwise> 7 meters wide , 2 meters high
[09:09] <brobostigon> ah.
[09:09] <knightwise> needs to be reinforced with iron rods and filled up with concrete
[09:09] <brobostigon> i see.
[09:10] <knightwise> and we live in a row of houses
[09:10] <knightwise> so everything needs to be carried THROUGH the house
[09:10]  * knightwise is getting some exercise
[09:10] <brobostigon> :(
[09:11] <knightwise> its shitty but it needs to be done
[09:11]  * knightwise listening to a good audiobook meanwhile ...
[09:11] <brobostigon> yep.
[09:11] <knightwise> so that helps to pass the time
[09:11] <brobostigon> :)
[09:55] <bigcalm> Sometimes I wish phone calls came with sub titles. Just had a call from a lady with a very thick Welsh accent. Think I got the gist of what she said :S
[10:09] <knightwise> lol
[12:02] <diddledan> morning
[12:21] <diddledan> grr @ adobe dropping support for linux flash

[13:25] <bigcalm> Anybody know much about children's rocking horses? I'm going to buy this off a friend but have no idea how much to offer her: https://www.dropbox.com/s/awce26xa6eq9kyz/IMG_20140717_173455_705.jpg
[13:26] <bigcalm> I'm having a bugger of a time finding something similar on the interwebs
[14:12] <mapps> argh
[14:13] <mapps> pesky power cable came out again
[14:15] <diddledan> don't you hate when you unexpectedly fall out?
[14:16] <mapps> yes!
[14:16] <mapps> its been dodgy for a while
[14:16] <mapps> have to wedge a remote under it
[19:01] <popey> evening
[19:25] <daftykins> what-o
[19:27] <daftykins> ok, test VM running 12.04.5 - run "hwe-support-status --verbose" and i get...
[19:27] <daftykins> "You are not running a system with a Hardware Enablement Stack. Your system is supported until April 2017."
[19:27] <daftykins> i was under the impression it's going to automatically get trusty's HWE
[19:28] <daftykins> it's certainly claimed that the above command is meant to provide instructions on how to install it if not
[19:28] <daftykins> (just for clarification here, i don't want to know HOW to do it... i'm curious that the above doesn't provide an answer)
[19:56] <popey> wat https://www.kickstarter.com/projects/mmv/console-os-dual-boot-android-remastered-for-the-pc
[20:13] <diddledan> popey: that's nuts
[20:32] <daftykins> "Most other processes on the system trace their origin ultimately to init; the exceptions are kernel processes, started by the kernel directly for managing internal operating system details."
[20:32] <daftykins> any thoughts on examples of kernel processes? :)
[20:32] <daftykins> might driver modules count?
[20:32] <daftykins> i suppose they aren't processes
[20:33] <popey> look for any process on your system owned by root starting with the letter k
[20:33] <popey> kworker, kthread, khelper, kdevtmpfs, kcryptd etc
[20:34] <daftykins> mm-hmm
[20:34] <daftykins> lots of kworker yeah
[20:38] <daftykins> ty sir
[20:50] <ali1234> modules definitely aren't separate processes
[20:51] <daftykins> *nod*
[20:51] <ali1234> kernel uses processes for long running worker tasks
[20:51] <daftykins> as soon as i wrote it i figured, well hey there's no 'nvidia' process ;)
[20:51] <ali1234> actually there is
[20:52] <daftykins> damn.
[20:52] <ali1234> modules can use one or more threads/processes
[20:52] <daftykins> a bit of research seems to say things like kworker handle kernel processing, such as interrupts, timers and I/O
[20:52] <ali1234> most don't though - most are really really simple
[20:52] <ali1234> but module code is no different to any other kernel code really
[20:53] <ali1234> yeah interrupts is a good example
[20:54] <daftykins> would that be in terms of IRQ handling?
[20:54] <ali1234> the interrupt itself fires in an interrupt context which means it interrupts absolutely everything else happening on the computer
[20:54] <daftykins> or a different kind of interrupt
[20:54] <ali1234> IRQs yes
[20:54] <ali1234> so you want to handle the interrupt as quickly as possible to prevent them from piling up, which is bad
[20:55] <daftykins> *nod*
[20:55] <ali1234> so what you do is collect the interrupt information and hand it over to a kworker thread to deal with then leave the interrupt context
[20:56] <ali1234> most modules do not need to directly deal with interrupts
[20:56] <ali1234> for example USB - all the interrupt stuff is andled in the host controller driver (uhci/ohci/ehci/xhci)
[20:57] <ali1234> the actual device drivers never see the details, they just get data streams
[20:57] <ali1234> so they don't need any threads usually
[20:59] <daftykins> might DMA IO differ from involving a kworker thread?
[20:59] <daftykins> i think that bypasses CPU involvement?
[20:59] <ali1234> correct, DMA bypasses the CPU... or more correctly it is "fire and forget"
[21:00] <ali1234> you say "copy from address x to address y length z" and it does it and signals back when it is done (probably by asserting an IRQ)
[21:01] <daftykins> i suspect that's why a lot of software graphically stalls when awaiting IO then, as there's no mechanism to say there's a problem until it's done
[21:02] <ali1234> that stems from a lack of threaded programming
[21:02] <daftykins> i say graphically, but that's probably just in the case of a blocking design
[21:02] <ali1234> yes
[21:02] <ali1234> that's not really down to the kernel, there are ways to avoid it if you program "correctly"
[21:02] <daftykins> mmm it's interesting discovering how the pieces come together more, i'm just doing this newbie edX course for fun :)
[21:03] <daftykins> it's referring to init and the older System V UNIX style of doing things
[21:03] <daftykins> though i understand ubuntu doesn't use runlevels and hasn't done for years (well, rather it only uses two in a way)
[21:03] <ali1234> it's still widely used despite what some people will tell you
[21:04] <daftykins> mmm the course material states that the conventions of System V are kept for compatibility
[21:04] <ali1234> busybox init is probably the most widely used init in the world :)
[21:05] <daftykins> for the swathes of embedded devices?
[21:05] <ali1234> right
[21:09] <daftykins> thanks :)
[21:09]  * daftykins continues with the boot process chapter
[21:46] <popey> ali1234: hows your racing game coming on?
[21:46] <ali1234> not bad. i just rewrote the track system again recently
[21:47] <ali1234> i'm using catmull-clark surfaces now, which means you can directly export an arbitrary surface from blender and that's the track
[21:47] <ali1234> however, this means that tracks can have arbitrary topology, which means there's no way to determine who is winning
[21:48] <popey> heh
[21:48] <ali1234> i also did a test with sound using soloud
[23:33]  * xnox canonical--
[23:34]  * xnox not-canonical++