| 00:40.52 | *** join/#edev s1axter (~s1axter@c-98-245-7-56.hsd1.co.comcast.net) |
| 00:57.30 | *** join/#edev djerome (~djerome@ip68-2-20-108.ph.ph.cox.net) |
| 01:07.49 | sunrunner20 | woa, holy netsplit |
| 01:08.06 | sunrunner20 | anyway, has anybody tried hacking Kingston Wi drive? |
| 01:17.49 | roxfan | http://www.kingston.com/en/support/technical/downloads?product=WID&filename=Wi-Drive_SDK2.4 |
| 01:18.12 | roxfan | sounds like linux |
| 01:19.29 | sunrunner20 | oh, I didn't think to look for a GPL kit |
| 01:24.08 | roxfan | first thing i'd do :) |
| 01:24.31 | SpeedEvil | Second thing is to find people bitching about the GPL status. |
| 01:32.34 | *** join/#edev War2 (~war2@230.Red-217-125-111.staticIP.rima-tde.net) |
| 01:32.40 | *** join/#edev War2 (~war2@unaffiliated/war2) |
| 02:05.55 | *** join/#edev prpplague (~prpplague@ppp-70-249-149-70.dsl.rcsntx.swbell.net) |
| 02:11.26 | *** join/#edev uwe_ (~uwe_@dslb-088-066-186-052.pools.arcor-ip.net) |
| 04:11.35 | *** join/#edev rsFF (~rsFF@bl15-3-7.dsl.telepac.pt) |
| 05:33.49 | *** join/#edev risca (~risca@wi-secure-434.cc.umanitoba.ca) |
| 06:24.55 | *** join/#edev clio (~andrej@85.159.109.222) |
| 07:40.13 | *** join/#edev factor (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net) |
| 08:20.29 | *** join/#edev thannoy (~aloiseau@231.39.0.109.rev.sfr.net) |
| 08:55.10 | *** join/#edev factor (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net) |
| 09:26.36 | *** join/#edev War2 (~war2@unaffiliated/war2) |
| 10:25.31 | *** join/#edev rsFF (~rsFF@Bl19-53-127.dsl.telepac.pt) |
| 11:10.10 | *** join/#edev lfelipe (~lfelipe@201.82.19.44) |
| 11:10.11 | *** join/#edev lfelipe (~lfelipe@enlightenment/developer/lfelipe) |
| 12:12.13 | *** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac) |
| 13:25.25 | *** join/#edev War2 (~war2@unaffiliated/war2) |
| 13:42.23 | *** join/#edev sunrunner20 (~jamesdoe@pool-71-123-194-101.dllstx.fios.verizon.net) |
| 14:01.56 | SpeedEvil | http://uk.farnell.com/raspberry-pi/raspbrry-chipset/chipset-raspberry-pi-model-b/dp/2081347?Ntt=2081347 - now - this is intersting - if true. |
| 14:02.05 | SpeedEvil | <PROTECTED> |
| 14:02.14 | SpeedEvil | If it comes with schematics. |
| 14:07.15 | *** join/#edev GPSFan (~kenm@64.92.145.112) |
| 14:07.16 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:10.48 | *** join/#edev GPSFan (~kenm@64.92.145.112) |
| 14:10.49 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:23.39 | *** join/#edev file` (~file@2001:4978:2d7:0:20c:29ff:feff:49c0) |
| 14:24.18 | *** join/#edev file (~file@asterisk/developer-and-muffin-lover/file) |
| 15:32.17 | *** join/#edev risca (~risca@wi-secure-434.cc.umanitoba.ca) |
| 15:38.47 | *** join/#edev prpplague (~danders@nat/ti/x-fiewmuneaevchegm) |
| 15:45.28 | prpplague | wmat: ping |
| 15:46.08 | av500 | prpplague: shush, people are trying to sleep |
| 15:46.14 | prpplague | hehe |
| 15:46.45 | sunrunner20 | lol |
| 15:47.05 | sunrunner20 | I plan to get a few raspberry pi's as soon as I have the cash |
| 15:47.26 | av500 | must save long? |
| 15:51.56 | *** join/#edev War2 (~war2@unaffiliated/war2) |
| 15:52.30 | sunrunner20 | they're like $26 each, but I'm dead broke |
| 15:52.51 | sunrunner20 | and need to concentrate on finding a job not (whats currently) screwing around |
| 15:55.24 | sunrunner20 | so if anybody knows an entry level IT position for a measly associates student I'd love to know |
| 16:10.43 | slashbeast | Anyone can confirm that 64bit busybox ignore umask when doing mkdir? |
| 16:10.45 | slashbeast | https://bugs.busybox.net/show_bug.cgi?id=4814 |
| 16:11.28 | av500 | sunrunner20: US I guess? |
| 16:11.43 | sunrunner20 | ? |
| 16:11.49 | av500 | your locale |
| 16:12.06 | sunrunner20 | yea, Right next to prpplague unless he's moved |
| 16:12.11 | sunrunner20 | (Dallas, Texas) |
| 16:12.18 | prpplague | is still in the dfw area |
| 16:12.19 | av500 | sunrunner20: intern at ti? |
| 16:12.34 | av500 | or, gofer for prpplague |
| 16:12.48 | prpplague | hehe |
| 16:12.56 | sunrunner20 | I woulden't mind |
| 16:13.00 | sunrunner20 | embedded is in the family |
| 16:13.01 | av500 | sold! |
| 16:14.04 | wmat | prpplague: pong |
| 16:15.19 | sunrunner20 | biggest issue is I'm currently doing an associates degree. Hopeing to do that in 3 semesters (normally 5 ish), get a job, then go back to UTD for a CS/CE degree |
| 16:16.01 | sunrunner20 | associates D/Q's me from a lot of internships |
| 16:23.26 | prpplague | is grumpy today |
| 16:23.52 | prpplague | my wife informs me this morning that my mother-in-law is coming to visit |
| 16:24.19 | av500 | rhianna? |
| 16:24.49 | prpplague | spent a good portion of the morning catering to my daughter who seems to have bad menstrual cramps |
| 16:25.04 | SpeedEvil | the 'chipset' from the Rpi - if real - may be interesting to play with. |
| 16:25.11 | av500 | chipset? |
| 16:25.13 | prpplague | got a call from the school and my son is failing social studies because he hasn't turned in his homework |
| 16:25.17 | av500 | north and souhtbridge? |
| 16:25.22 | SpeedEvil | http://uk.farnell.com/raspberry-pi/raspbrry-chipset/chipset-raspberry-pi-model-b/dp/2081347?Ntt=2081347 |
| 16:25.24 | prpplague | my board bringup is giving me trouble |
| 16:25.27 | SpeedEvil | Claims to be the chipset. |
| 16:25.32 | SpeedEvil | If this is real, I'm unsure. |
| 16:25.38 | prpplague | and _SOMEONE_ over in another cube is playing music loudly |
| 16:26.00 | prpplague | av500: hehe, no Fenty's at house atm |
| 16:26.04 | SpeedEvil | av500: More like SoC and POP RAM, probably. |
| 16:26.33 | av500 | easy ot hand solder.... |
| 16:26.34 | SpeedEvil | av500: Maybe USB hub and ethernet chip, depending on what they're doing. If it isn't just an error. |
| 16:26.35 | av500 | to |
| 16:26.41 | SpeedEvil | av500: No. :) |
| 16:27.02 | SpeedEvil | av500: But maybe fun to play around with anyway if you can ghetto up a reflow oven that can do the right profile. |
| 16:27.10 | SpeedEvil | and get stencil and PCB |
| 16:27.41 | SpeedEvil | Stencils (of some form) are now 'free' to me - due to my brother getting a laser printer. |
| 16:27.49 | SpeedEvil | s/printe/cutte/ |
| 16:28.08 | sunrunner20 | o.0 since when does edev have a bog |
| 16:28.13 | sunrunner20 | *bot |
| 16:28.28 | Mirell | It's had a bot for years now |
| 16:29.26 | sunrunner20 | Its been years since I frequented |
| 16:30.19 | prpplague | crap |
| 16:30.27 | prpplague | someone left an audio test running |
| 16:33.18 | *** join/#edev sndcrb (~sndcrb@dsl-hkibrasgw2-fef8de00-126.dhcp.inet.fi) |
| 16:37.26 | *** join/#edev risca (~risca@wi-secure-434.cc.umanitoba.ca) |
| 16:43.58 | MonMotha | there's essentially no way you're going to be able to do one of those micro pitch POP setups with a toaster oven |
| 16:44.05 | MonMotha | most CMs don't have the ability to do it |
| 16:44.19 | av500 | pop is pretty easy these days |
| 16:44.19 | MonMotha | the ball pitch is like 0.4mm, and you have to manage to stack the packages just right |
| 16:44.50 | MonMotha | really? all the CMs I've talked with are like "uh, don't do that if you can avoid it" (the implication being "you won't like how much it costs") |
| 16:45.06 | MonMotha | the pitch alone on those chips makes them a pain |
| 16:45.15 | av500 | yes, 0.4 is a pain |
| 16:47.05 | MonMotha | I just really don't see somebody managed to reflow one successfully in a toaster oven |
| 16:47.06 | av500 | it makes the pcb expensive because you need stuff like buried vias |
| 16:47.15 | MonMotha | and yeah, the PCB is going to be ugly in low volume |
| 16:47.20 | av500 | yes |
| 16:47.45 | av500 | but you can always order it in china :) |
| 16:48.17 | SpeedEvil | MonMotha: Getting the alignment right is tricky, but some simple ovens can do a similar profile. |
| 16:48.28 | SpeedEvil | And the PCB is of course going to be expensive. |
| 16:48.48 | MonMotha | yeah, I'm not overly concerned with the temperature profile. It's more alignment and handling. |
| 16:48.54 | SpeedEvil | I came across an appnote from TI - the POP mounting looks not-too-bad, especially if you do nitrogen. |
| 16:49.27 | av500 | prefers N2O |
| 17:07.48 | *** join/#edev factor (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net) |
| 17:38.34 | *** join/#edev CosmicPenguin (~nobody@soa.codeaurora.org) |
| 17:38.34 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 18:03.59 | *** join/#edev prpplague (~danders@192.91.66.186) |
| 18:45.28 | *** join/#edev dijenerate (~dijenerat@63.143.73.34) |
| 19:04.48 | *** join/#edev landley (~landley@140.242.26.2) |
| 19:21.39 | *** join/#edev dijenerate (~dijenerat@173.225.251.175) |
| 19:36.06 | *** join/#edev samxor (simoirc@206.217.201.48) |
| 19:36.07 | landley | didn't I blather about this here last week? :) |
| 19:36.09 | landley | http://arstechnica.com/business/news/2012/02/how-red-hat-killed-its-core-productand-became-a-billion-dollar-business.ars |
| 19:57.42 | *** join/#edev factor (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net) |
| 20:01.41 | sunrunner20 | I have to question the wisdom of trying to encroach upon vmware |
| 20:05.41 | sunrunner20 | I don't know so much about the server market, but I know desktop wise since Player got VM creation I don't know anybody who uses anything else |
| 20:10.58 | calculus | I have used virtualbox (now Oracle, formerly Sun, formerly someone else) |
| 20:15.53 | landley | sunrunner20: at work I use virtualbox. In the open source world, I use kvm or qemu depending on the need. The rackmount/cloud guys use container-based solutions (openvz switching to lxc) becausethey're an order of magnitude more efficient at packing lots of images onto a single server. |
| 20:16.14 | landley | This leaves vmware and xen fighting for "I don't care" pointy-haired cobol territory. |
| 20:16.22 | sunrunner20 | ah |
| 20:16.45 | landley | I should write up the "container vs virtual machine" spiel sometime. |
| 20:17.11 | landley | A VM is emulating hardware and booting a separate OS on it. A container is basically chroot on steroids, running lots of processes under one kernel and just collecting them together into groups. |
| 20:17.51 | landley | In the container world, if I have 37 instances of bash running, they can share the underlying physical pages, even if they belong to different users _or_ different containers. |
| 20:18.07 | landley | In the vm world, I can't even share disk cache between instances. |
| 20:18.10 | kergoth | I really nee dto sit down and play with containers more |
| 20:18.50 | sunrunner20 | why virtualbox over player? |
| 20:19.26 | landley | Virtual machine started with full isolation and then try towork their way towards efficiency by punching holes in their emulation. Things like virtual devices that pass data more efficiently than emulating real hardware, tickless kernels, balloon drivers, and so on. |
| 20:19.46 | landley | But the end point of virtual machines is reinventing the microkernel. You still spend all your time passing messages to yourself. |
| 20:20.22 | kergoth | landley: are there userland container tools that don't suck yet? |
| 20:20.27 | landley | Containers start with Unix's ability to siolate processes from each other (users and chroots) and add a new concept, a "container", which gives them a new way to group/isolate the suckers. |
| 20:20.29 | kergoth | lxc looked rather unpleasant |
| 20:21.07 | landley | kergoth: the openvz stuff has been going since 1999. The problem is it's all based on a giant out of tree fork, none of which is going upstream because when they finally got around to submitting it Linus vetoed the design approach. (Adding buckets of syscalls to implement the new features). |
| 20:21.58 | landley | Instead he said "add a new clone flag to control separate namespaces, make a new filesystem to handle resource allocation, and the rest is details like -o newinstance flag for filesystems like /proc and devpts. |
| 20:22.30 | landley | kergoth: lxc is IBM's attempt at making a userspace control for the new container infrastructure in the kernel. It's about what you'd expect from IBM: a mainframe boondoggle. |
| 20:22.36 | kergoth | interesting |
| 20:22.36 | kergoth | nods |
| 20:22.40 | landley | The kernel infrastructure's pretty nice. I need to do a UI thing for it in toybox. |
| 20:22.50 | landley | (I added unshare already. I need to do the cgroups and init bits...) |
| 20:23.02 | landley | tries to dig up old notes he wrote on this... |
| 20:23.24 | landley | Well, I found http://landley.net/lxc/ anyway... |
| 20:23.42 | landley | Nah, I need to do a proper writeup on this... |
| 20:25.28 | calculus | I saw landley representing containers and migrating them from one box to another with no downtime :) |
| 20:25.43 | landley | calculus: ah, you were at Scale a year ago? :) |
| 20:26.07 | calculus | landley: yep, since Scale 4x |
| 20:26.09 | landley | Kir Kolyshkin (the openvz kernel maintainer) and I ran a booth at Scale when I was doing a contract for Parallels. |
| 20:26.49 | landley | Parallels employs most of the OpenVZ developers. They have a VM thing for MacOS but most of the engineers think containers are the future. |
| 20:27.22 | landley | They're working to migrate their userspace tools on top of the vanilla container stuff that's going into the kernel, it's just that working out full isolation for containers is about as hard as getting "non-root user can't escalate to root without the password" worked out properly was. |
| 20:27.56 | landley | Random bug du jour I remember: /proc/sys/vm/drop_caches lets you free all the CPU cache. If you do it in a container, it shouldn't free EVERYBODY'S CPU cache, but the system doesn't really track cache on a per-container basis, so... |
| 20:28.11 | landley | Most likely it should be ignored from non-root context... |
| 20:28.31 | landley | They're adding all sorts of resource tracking stuff so they can enforce the various low water marks and high water marks for each container. |
| 20:28.50 | landley | This is basically "ulimit, but for an aribitrary collection of processes (and all child processes they spawn) instead of for a user." |
| 20:29.21 | landley | So you can say "never let the container use more than X megabytes, but reserve X megabytes for it that the rest of the system can't use". That's the low water marks and high water marks. |
| 20:29.21 | SpeedEvil | Interesting. |
| 20:29.48 | landley | was working on a paper for this for OLS a year ago, but the OLS guys never got their act together to tell me where to SEND the paper so it never went beyond the "outline and some notes" stage. |
| 20:30.52 | landley | So with containers you can tell the system "all the processes in this container, running flat out, can never use more than 50% of the CPU, 60% of the memory, and 30% of the I/O bandwidth. But no matter how busy the _rest_ of the system gets, guarantee 20% of each for this container". |
| 20:31.20 | landley | The high water marks say "leave the rest of the system idle if necessary". The low water marks mean "let the rest of the system starve". |
| 20:31.32 | landley | You can overcommit the high water marks all you like, 100% for everybody is actually the default. |
| 20:31.46 | landley | But if you try to overcommit the low water marks, it'll fail with an error trying to set 'em. |
| 20:41.55 | *** join/#edev factor (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net) |
| 21:08.31 | *** join/#edev flo_lap (~fuchs@Maemo/community/contributor/florian) |
| 21:51.17 | *** join/#edev risca (~risca@wi-secure-434.cc.umanitoba.ca) |
| 22:30.00 | *** join/#edev cyronin (~travis@ip68-10-232-109.hr.hr.cox.net) |
| 22:42.48 | *** join/#edev lfelipe (~lfelipe@186.214.74.77) |
| 22:42.49 | *** join/#edev lfelipe (~lfelipe@enlightenment/developer/lfelipe) |
| 22:54.51 | *** part/#edev flyback (~flyback@c-98-219-134-156.hsd1.pa.comcast.net) |
| 22:57.24 | *** join/#edev bgamari (~ben@physicsnat56.physics.umass.edu) |
| 23:22.32 | sunrunner20 | oh, texas drivers license's have some fancy new anticounterfiting features |
| 23:23.44 | SpeedEvil | A picture? |
| 23:24.06 | sunrunner20 | of the features? |
| 23:24.14 | alex4nder | laf. |
| 23:26.09 | sunrunner20 | well, DOB is embossed into the plastic. , DOB on the paper is printed with the dual color ink (can't remember what its called), holo foil all over the place |
| 23:26.22 | sunrunner20 | and an outline of texas is punched or laser cut through the paper |
| 23:34.21 | g1powermac | think there's less anti counterfeit stuff in US passports. . . |
| 23:40.08 | sunrunner20 | theyre might be a cryptogrpahic signarure on the RFID |
| 23:44.43 | g1powermac | I remember looking at my parent's license when I was young, it didn't even have a picture and if you changed address, the state would just send you a sticker to put over the wrong address |
| 23:45.17 | sunrunner20 | heh |
| 23:45.32 | g1powermac | and that was probably only 15 years ago |
| 23:45.45 | sunrunner20 | Just researched the pasport rfid. I couldn't find anythign about a cryptographic signature, but the encryption was quickly broken |
| 23:46.33 | g1powermac | ahh |
| 23:47.00 | sunrunner20 | which makes me wonder what idiot designed it, unless the breakers used a sidechannel attack or something |
| 23:47.54 | g1powermac | has there been a any well known encryption system that hasn't been broken via some method? |
| 23:49.13 | sunrunner20 | I think TLS has only been broken with MITM or code injection |
| 23:50.29 | g1powermac | I personally would love to see a system designed to not need encryption at all |
| 23:51.12 | SpeedEvil | Umm - what? |
| 23:52.20 | sunrunner20 | SpeedEvil: ? |
| 23:52.35 | SpeedEvil | I'm curious how you'd do it without crypto |
| 23:53.02 | sunrunner20 | oh |
| 23:53.12 | sunrunner20 | I thought I had missed something with my TLS statement |
| 23:53.13 | g1powermac | SpeedEvil, no idea, the idea just popped up in my mind without much thought about it ;-) |
| 23:53.43 | sunrunner20 | I think entanglement could theoretically be used |
| 23:53.51 | sunrunner20 | but currently its rubbish |
| 23:54.40 | SpeedEvil | Interesting idea. |
| 23:54.51 | SpeedEvil | A kitten and a ball of wool as a one-time-knot generator. |
| 23:55.22 | sunrunner20 | isn't there an inherent problem when the key is smaller than the data though? |
| 23:58.14 | SpeedEvil | Not really. |
| 23:58.39 | sunrunner20 | oh, i meant on hashes |
| 23:58.52 | SpeedEvil | The issue resolves to 'If you can't crack it if you convert the entire mass of the earth into nanobots crunching the key until the sun dies, it's probably secure' |
| 23:59.33 | sunrunner20 | if hash < data then there WILL be collisions |
| 23:59.42 | sunrunner20 | unless my math is off |