Sunday, 2020-01-12

*** tpb has joined #symbiflow00:00
*** jjjaaaccckkk has joined #symbiflow01:24
jjjaaaccckkkfanyone have Mathias Lasser's full slide deck from his presentation at 34C3 about reverse engineering the xilinx series 7?01:26
*** alexhw has quit IRC02:17
az0rejjjaaaccckkk: https://media.ccc.de/v/34c3-9237-reverse_engineering_fpgas doesn't work for you?02:19
tpbTitle: media.ccc.de - Reverse engineering FPGAs (at media.ccc.de)02:19
az0reAh, just saw your message in #yosys... sorry, can't help02:21
*** alexhw has joined #symbiflow02:22
jjjaaaccckkkno worries, yeah I was hoping someone has all the slides from his presentation. In the video he seemed glad to share them with whoever was interested02:28
jjjaaaccckkkOr anyone know Mathias and can ask if he is open to sharing them?02:30
*** citypw has joined #symbiflow03:08
*** Bertl_oO is now known as Bertl_zZ03:29
*** OmniMancer has joined #symbiflow03:51
*** _whitelogger has quit IRC04:19
*** _whitelogger has joined #symbiflow04:21
*** daveshah has quit IRC04:36
*** _florent_ has quit IRC04:36
*** bubble_buster has quit IRC04:37
*** daveshah has joined #symbiflow04:37
*** Thalheim has quit IRC04:38
*** bubble_buster has joined #symbiflow04:39
*** Thalheim has joined #symbiflow04:39
*** _florent_ has joined #symbiflow04:50
*** phire has quit IRC04:55
*** _whitelogger has quit IRC05:37
*** _whitelogger has joined #symbiflow05:39
*** proteus-guy has quit IRC06:09
*** proteus-guy has joined #symbiflow06:10
*** _whitelogger has quit IRC08:37
*** _whitelogger has joined #symbiflow08:39
*** rvalles_ has quit IRC09:34
*** rvalles_ has joined #symbiflow09:48
*** adjtm_ has joined #symbiflow10:14
*** adjtm has quit IRC10:16
*** Bertl_zZ is now known as Bertl10:36
*** citypw has quit IRC12:45
*** Bertl is now known as Bertl_oO14:55
*** OmniMancer has quit IRC15:06
* bunnie[m] sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/UIkSZnmdAnMMCUPdwkTgLwlA >15:13
*** _whitelogger has quit IRC15:22
*** _whitelogger has joined #symbiflow15:24
ZirconiumXbunnie[m]: the Intel bitstream is even stranger15:44
bunnie[m]yow.15:58
*** mario_h has joined #symbiflow17:56
mithrojjjaaaccckkk: I attended and he didn't have any slides and pretty much no real details19:15
*** phire has joined #symbiflow19:17
*** mario_h has quit IRC19:43
jjjaaaccckkkThanks, mithro21:06
mithroIt is in no way connected to prjxray21:19
jjjaaaccckkkYep I knew that21:28
az0reWhat is the difference between prjxray and his work?21:29
mithroaz0re: his work doesn't exist as far as I can tell, prjxray is mostly just missing DSP + Serdes21:37
az0reGotcha, thanks21:46
mithroprjxray is at https://github.com/SymbiFlow/prjxray and prjxray.rtfd.io21:49
tpbTitle: GitHub - SymbiFlow/prjxray: Documenting the Xilinx 7-series bit-stream format. (at github.com)21:49
jjjaaaccckkkI am curious, though, do you understand the technique Mathias talked about? It sounds like he was implementing Vivado example projects and reading each pixel in the device view image to see what bits were set?22:25
az0remithro: Cool work.  Thanks for doing it!22:27
az0reI haven't really dug into it yet, but my impression is that the process is roughly like:22:27
az0re1. Read documentation, get an idea of what resources exist on the FPGA family22:28
mithrojjjaaaccckkk: You generate a lot of bitstreams and then look for cross correlation between bits and features22:28
mithrojjjaaaccckkk: There is some info in the docs above22:28
az0re2. Write fuzzer generator scripts to instantiate those IP features in uniquely identifiable ways22:28
az0re3. Do the full design flow in the Xilinx tools, observe changes in the bitstream22:28
az0re4. Correlate unique identifiers (or uniquely identifying behaviors?) with bits in the bitstream22:29
az0reIs that about right?22:29
-_whitenotifier-5- [sphinx_materialdesign_theme] mithro opened issue #1: script_files in the theme is deprecated - https://git.io/JvfbR23:17
*** jjjaaaccckkk has quit IRC23:31
mithroaz0re: pretty much23:43

Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!