| 00:17.20 | *** join/#edev flyback (~whatthe@c-98-219-134-156.hsd1.pa.comcast.net) |
| 00:17.39 | *** part/#edev flyback (~whatthe@c-98-219-134-156.hsd1.pa.comcast.net) |
| 03:38.35 | *** join/#edev JoeLlama (~snork@unaffiliated/joellama) |
| 03:54.44 | *** join/#edev juni_kyu (~chatzilla@96.18.213.235) |
| 04:03.03 | *** join/#edev miltonm (~miltonm@71.22.127.106) |
| 04:09.14 | *** join/#edev miltonm (~miltonm@71.22.127.106) |
| 05:18.03 | *** join/#edev mbuf (~user@nat/redhat/x-wefzoawfssmpeqwv) |
| 06:00.17 | *** join/#edev gvandenbosch (~gvandenbo@86.81.67.223) |
| 07:11.47 | *** join/#edev risca (~risca@nl105-253-30.student.uu.se) |
| 07:39.44 | *** join/#edev likewise (~likewise@42-81-ftth.onsneteindhoven.nl) |
| 07:53.24 | *** join/#edev MonMotha (~monmotha@108-71-210-59.lightspeed.iplsin.sbcglobal.net) |
| 08:48.20 | *** join/#edev florian_kc (~fuchs@port-217-146-132-69.static.qsc.de) |
| 08:48.20 | *** join/#edev florian_kc (~fuchs@Maemo/community/contributor/florian) |
| 11:41.24 | *** join/#edev gustavoz (~gustavoz@host20.186-109-4.telecom.net.ar) |
| 12:42.10 | *** join/#edev madsy (~madsy@188.113.83.44) |
| 12:42.10 | *** join/#edev madsy (~madsy@fu/coder/madsy) |
| 13:00.01 | *** join/#edev GPSFan (~kenm@64.92.145.112) |
| 13:00.01 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:05.58 | *** join/#edev Rubix (~Rubix@75.144.143.69) |
| 13:19.11 | *** join/#edev Rubix (~Rubix@75.144.143.69) |
| 13:24.22 | *** join/#edev ccole (~cole@coledd.com) |
| 13:31.28 | *** join/#edev avinashhm (~avinash-h@192.91.66.189) |
| 13:32.39 | *** join/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 13:33.45 | *** join/#edev risca (~risca@217.174.82.198) |
| 13:36.34 | *** part/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 14:32.00 | *** join/#edev sndcrb (~sndcrb@dsl-hkibrasgw2-fef8de00-126.dhcp.inet.fi) |
| 15:13.00 | *** join/#edev risca (~risca@nl105-253-30.student.uu.se) |
| 15:17.17 | *** join/#edev dijenerate (~dijenerat@64.210.44.28) |
| 15:21.44 | *** join/#edev CosmicPenguin (~nobody@soa.codeaurora.org) |
| 15:30.05 | *** join/#edev prpplague (~danders@nat/ti/x-depfscqumwtgbero) |
| 15:37.39 | *** join/#edev dijenerate (~dijenerat@204.212.240.167) |
| 15:46.57 | *** join/#edev Mirell (~mmiller@maetel.mirell.org) |
| 15:46.57 | *** mode/#edev [+o Mirell] by ChanServ |
| 15:53.23 | *** mode/#edev [+o prpplague] by ChanServ |
| 15:59.24 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 16:03.41 | *** join/#edev jayabharath (~jayabhara@nat/ti/x-wwxsoclyfctumwcf) |
| 16:16.28 | *** join/#edev madsy (~madsy@188.113.83.44) |
| 16:16.28 | *** join/#edev madsy (~madsy@fu/coder/madsy) |
| 16:56.23 | *** join/#edev sjs205 (~sjs205@host86-182-135-222.range86-182.btcentralplus.com) |
| 17:05.56 | *** join/#edev vanner (~vanner@195.91.241.19) |
| 17:12.45 | *** join/#edev gvandenbosch (~gvandenbo@86.81.67.223) |
| 17:12.45 | *** join/#edev Zygo (startkeylo@startkeylogger.hungrycats.org) |
| 17:12.45 | *** join/#edev jevin (~jevin@jevinskie-1-pt.tunnel.tserv9.chi1.ipv6.he.net) |
| 17:12.45 | *** join/#edev devilsadvocate (~quassel@chintal-1-pt.tunnel.tserv10.par1.ipv6.he.net) |
| 17:12.45 | *** join/#edev htns (~htns@61.6.64.6) |
| 17:23.13 | *** join/#edev bgamari (~ben@pool-72-79-221-71.spfdma.east.verizon.net) |
| 18:02.08 | *** join/#edev Rubix (~Rubix@75.144.143.69) |
| 19:05.44 | *** join/#edev florian (~fuchs@sign-4d094a2c.pool.mediaWays.net) |
| 19:05.44 | *** join/#edev florian (~fuchs@Maemo/community/contributor/florian) |
| 19:30.19 | *** join/#edev sjs205 (~sjs205@host86-150-138-126.range86-150.btcentralplus.com) |
| 20:09.05 | *** join/#edev Rubix (~Rubix@75.144.143.69) |
| 20:18.40 | *** join/#edev NightMonkey (debian-tor@pdpc/supporter/professional/nightmonkey) |
| 20:30.55 | *** join/#edev orbarron (~orb@nat/ti/x-voesnfniezlezppk) |
| 20:52.27 | *** join/#edev NightMonkey (debian-tor@pdpc/supporter/professional/nightmonkey) |
| 21:05.06 | *** join/#edev CosmicPenguin (~nobody@soa.codeaurora.org) |
| 21:13.55 | CosmicPenguin | yo, lazyweb |
| 21:14.09 | CosmicPenguin | is there a compiler attribute that says "this cannot be null" |
| 21:14.16 | CosmicPenguin | like for a const struct of function pointers? |
| 21:16.10 | *** join/#edev Rubix (~Rubix@75.144.143.69) |
| 21:22.35 | prpplague | CosmicPenguin: landley would be a good person to ask on that |
| 21:27.32 | *** join/#edev edboogie2k (~Edward_Ki@pool-173-77-212-165.nycmny.fios.verizon.net) |
| 21:27.55 | edboogie2k | hi guys. I ust finished writing my first bootloader, and wanted to know, should I use CRC32 check or MD5 to verify a firmware file before reflashing my microcontroller? |
| 21:28.03 | edboogie2k | oh and I also plan to use AES256 to decrypt my firmware... |
| 21:28.32 | edboogie2k | I am using a Cortex M3 ARM. NXP LPC1768 |
| 21:28.38 | SpeedEvil | CRC is fine to check for accidental corruption |
| 21:29.00 | SpeedEvil | decrypt the firmware, then check the checksum |
| 21:30.13 | edboogie2k | ok, so MD5 i guess is overkill |
| 21:30.25 | edboogie2k | i would like my total footprint of my bootloader to be small |
| 21:30.56 | SpeedEvil | There is no real need to do md5 on top of AES |
| 21:30.59 | roxfan | aes tables can get pretty big |
| 21:31.02 | SpeedEvil | AES + CRC is a secure hash |
| 21:31.09 | roxfan | you sure you need aes? |
| 21:31.33 | SpeedEvil | tries to recall if he's implemented AES in asm |
| 21:32.05 | edboogie2k | well i would like people to not be able to put their own hacked firmware on my device |
| 21:32.08 | SpeedEvil | No, it was MD5 I think |
| 21:32.09 | edboogie2k | and protect my IP |
| 21:32.29 | edboogie2k | is that not a good reason to use AES in my bootloader? |
| 21:32.33 | alex4nder | well using symmetric key crypto for that is probably not the best idea |
| 21:32.48 | SpeedEvil | I for one would often like to be able to pay a modest sum to unlock it an dbe able to put my own firmware on stuff. |
| 21:33.04 | CosmicPenguin | prpplague: according to TFM, the answer is no |
| 21:33.18 | edboogie2k | alex4nder: why is that? i don't know anything about crypto |
| 21:33.52 | alex4nder | edboogie2k: do you have a copy of Applied Cryptography by Schneier? |
| 21:33.55 | SpeedEvil | Public/private key crypto means that if they get the key from hacking your code somehow, they can't use it to make new images. |
| 21:33.59 | SpeedEvil | Indeed. |
| 21:34.07 | SpeedEvil | Get applied cryptography. |
| 21:34.15 | SpeedEvil | Much of this stuff is fairly complex. |
| 21:34.18 | alex4nder | SpeedEvil: exactly. |
| 21:34.28 | SpeedEvil | And needs more than a brief read over the manpage of the crypto function. |
| 21:35.00 | SpeedEvil | This is the problem of the Playstation people - they made fundamental errors in the crypto design. |
| 21:35.06 | SpeedEvil | Well - one problem. |
| 21:35.18 | alex4nder | the problem with crypto, is that everyone thinks they understand it, and then someone shows the world that you don't.. like the PS3 |
| 21:35.24 | alex4nder | haha |
| 21:35.34 | roxfan | why do you want to stop people from loading their own firmware? why do you care what they do with the device after you have your money? |
| 21:35.53 | alex4nder | roxfan: because they think it somehow helps protect the value of the product. |
| 21:36.00 | alex4nder | like, against the chinese |
| 21:36.03 | edboogie2k | uhm, i don't want to give them access to additional functionality they didn't pay for |
| 21:36.15 | roxfan | ahh |
| 21:36.17 | edboogie2k | well thats really not the reason |
| 21:36.22 | edboogie2k | but that would be a good reason |
| 21:36.24 | edboogie2k | lol |
| 21:36.27 | roxfan | so you want to sell one hw for different prices? |
| 21:36.28 | alex4nder | except it never works. |
| 21:36.45 | edboogie2k | i actually just want to protect my code |
| 21:37.00 | edboogie2k | really, if i keep the key in the protected flash |
| 21:37.12 | edboogie2k | people can still hack my code and read the plain binary? |
| 21:37.38 | alex4nder | edboogie2k: answer this: why do you think someone who is going to go to the trouble of pirating your value-added features, isn't going to figure out where your keys are? |
| 21:38.00 | edboogie2k | well my keys are in the protected area of the flash of my ARM |
| 21:38.08 | edboogie2k | but it costs a lot of money to hack that area |
| 21:38.14 | edboogie2k | way too much for any small company |
| 21:38.21 | roxfan | lpc1768 doesn't really have a "protected area" |
| 21:38.24 | SpeedEvil | edboogie2k: The stm has bits you can set, so that they can only wipe your code, not read it. |
| 21:38.26 | roxfan | it only has CRP |
| 21:38.30 | SpeedEvil | How secure these are, I'm unsure. |
| 21:38.49 | edboogie2k | roxfan: it doesnt? CRP is weak? i am new at edev, please tell me what you know about it.. |
| 21:39.14 | roxfan | crp is checked in the bootrom |
| 21:39.47 | roxfan | i've heard some people were able to stop the debugger at the bootrom entrypoint |
| 21:39.53 | roxfan | (so before crp is checked) |
| 21:40.16 | roxfan | not sure if that's actually possible but there might be some window of opportunity |
| 21:40.51 | edboogie2k | well assuming it is secure, isn't my scheme good? |
| 21:41.16 | alex4nder | how much does your product cost? |
| 21:41.24 | edboogie2k | $800 |
| 21:41.29 | edboogie2k | not too much |
| 21:41.43 | alex4nder | how many are you going to sell a year? |
| 21:41.58 | edboogie2k | i'm not sure yet, but i am hoping for 1000 first year |
| 21:42.05 | alex4nder | and how boutique is your customer? |
| 21:42.30 | edboogie2k | it is pretty boutique, but there are already 4 companies that make old designs that have been around for 10+ years |
| 21:43.52 | alex4nder | I dunno, I find the whole practice kind of pointless.. if someone wants it, there's a Ukranian with IDAPro, a JTAG/BDM interface, and plenty of time, waiting to help them. |
| 21:43.57 | alex4nder | +i |
| 21:44.40 | edboogie2k | lol.. how about potting with a high heat epoxy? |
| 21:44.58 | edboogie2k | i'm sure to get tons of hate on this topic |
| 21:45.04 | alex4nder | but I would suggest if you're going through the trouble of protecting your software, read Applied Crypto, and using a public/private key pair to sign firmware rather than just relying on one key. |
| 21:45.44 | SpeedEvil | In this app - is the private key really the public one, and vice-versa? |
| 21:45.51 | SpeedEvil | As you're assuming that the pirvate key can leak |
| 21:45.56 | alex4nder | they're really 'the same' |
| 21:46.03 | alex4nder | depends on which direction you're looking |
| 21:46.05 | SpeedEvil | yeah |
| 21:46.08 | SpeedEvil | I suppose |
| 21:46.36 | alex4nder | edboogie2k: if you're lucky to have people trying to steal your product, I think piracy will be the least of your worries. |
| 21:47.00 | alex4nder | unless you've got some new algorithm you're worried your competitors are going to pay to have disassembled. |
| 21:47.19 | alex4nder | and even then, that's good news. it means there's a market. |
| 21:48.18 | SpeedEvil | Also - if you are doing this sort of stuff - leave out the code, don't just enable it or disable it. So they at least need to buy one expensive version to copy. |
| 21:48.26 | SpeedEvil | In addition. |
| 21:48.40 | SpeedEvil | If you have seperate keys per device, that makes it lots more secure in some aspects. |
| 21:48.48 | SpeedEvil | But it complicates your maintainance. |
| 21:51.26 | alex4nder | SpeedEvil: yah, and that comes back to the style of customer |
| 21:51.46 | alex4nder | if it's a boutique product, you can maintain a tighter relationship, just don't apply draconian rules. |
| 21:52.20 | edboogie2k | ahh you guys mean like a key to the firmware that is based on the unit's serial number? something like that? where each key will be different per unit? |
| 21:53.31 | SpeedEvil | yes |
| 21:53.51 | SpeedEvil | So it's not possible to use a firmware from one device on another to upgrade it. |
| 21:54.16 | edboogie2k | thats clever... you could charge for firmware updates |
| 21:54.21 | edboogie2k | and premium features |
| 21:56.19 | alex4nder | edboogie2k: do you have The Four Steps to the Epiphany? |
| 21:57.10 | edboogie2k | no whats that? |
| 21:57.37 | edboogie2k | ohh its a business book |
| 21:58.56 | edboogie2k | you guys gave me 2 new books to put on my amazon list |
| 22:00.35 | edboogie2k | alex4nder: that book has 5 solid stars on amazon with 80 reviews |
| 22:00.44 | edboogie2k | wow must be the holy grail of buisness books |
| 22:02.03 | alex4nder | it's pretty good |
| 22:19.15 | *** join/#edev flyback (~whatthe@c-98-219-134-156.hsd1.pa.comcast.net) |
| 22:19.34 | *** part/#edev flyback (~whatthe@c-98-219-134-156.hsd1.pa.comcast.net) |
| 22:34.57 | *** join/#edev miltonm (~miltonm@71.22.127.106) |
| 22:35.36 | *** join/#edev Rubix (~Rubix@75.144.143.69) |
| 22:46.25 | *** join/#edev Rubix_ (~Rubix@75.144.143.69) |
| 22:54.01 | *** join/#edev Rubix (~Rubix@75.144.143.69) |
| 23:33.11 | *** join/#edev SpeedEvil1 (~user@tor/regular/SpeedEvil) |
| 23:40.41 | *** join/#edev prpplague (~prpplague@ppp-70-242-115-91.dsl.rcsntx.swbell.net) |
| 23:46.41 | *** mode/#edev [+o prpplague] by ChanServ |