<kc8apf> wire protocol format is in the datasheet. Looks like data is just stored in an external PROM. Can easily use a SPI flash or similar.
<cr1901_modern> Maybe that's what I confused w/ the bitstream format
<kc8apf> interestingly, #bits per frame and #frames increases as you move up the devices in the model line
<kc8apf> in their modern parts, only #frames increases. #bits per frame is constant in a series
<kc8apf> interesting to see a note about battery backup solutions to maintain the configuration
<kc8apf> also how to reconfigure during operation
<kc8apf> most of the concepts are the same as their modern devices
<cr1901_modern> 'cept for the JTAG part (meaning I'd have to program an SPI flash or whatever it expects directly)?
<kc8apf> yeah, the don't have the concept of multiple paths for programming
<kc8apf> you need to use whatever mode is selected by straps
<kc8apf> peripheral serial mode seems like you could program from a SPI master though
<cr1901_modern> I should buy one on Ebay and play w/ it when I have a free moment & I feel like it
<cr1901_modern> I imagine trying to get one of these things to work isn't actually going to be fun
<kc8apf> looks like slave mode is the easiest option. Could easily do that from a SPI interface.
* cr1901_modern nods
<cr1901_modern> I haven't actually looked in a while
GenTooMan has joined ##openfpga
<rqou> hrm, after fuzzing the mux bits a bit they don't _disagree_ with my hypothesis
<rqou> but it doesn't clearly support it either
<rqou> at this point i'm probably blocked on actually doing some exhaustive fuzzing rather than guessing
unixb0y_ has joined ##openfpga
unixb0y has quit [Ping timeout: 265 seconds]
unixb0y_ is now known as unixb0y
digshadow has quit [Ping timeout: 252 seconds]
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
digshadow has joined ##openfpga
azonenberg has joined ##openfpga
azonenberg has quit [Ping timeout: 240 seconds]
<digshadow> cr1901_modern: going to create an xc2064 repo
<digshadow> I'll let you know if I do anything meaningful with it
GenTooMan has quit [Quit: Leaving]
digshadow has quit [Ping timeout: 264 seconds]
<cr1901_modern> digshadow-c: Cool, tyvm.
<azonenberg> kc8apf: most of the modern parts are probably just replicas of the same basic tile
<azonenberg> the older parts were effectively a new asic for each chip
<kc8apf> azonenberg: sort of
<azonenberg> starting with 7 series the ASMBL architecture reuses a lot more of the physical layout and config stuff
<azonenberg> its much more orthogonal
<azonenberg> So they probably dont want the config structure to be different between chips
<azonenberg> it would not surprise me if in the not-too-distant future, you can create a partial bitstream that can be used as a hard IP core
<kc8apf> 7 series definitely has a regularity that is exposed in the frame addressing
<azonenberg> and will drop into e.g. any ultrascale+ part at any position aligned on a N-CLB boundary
<azonenberg> i.e. not locked to a given device density
<azonenberg> not sure how possible that last bit would be, given the varying mix of different column types
<azonenberg> but it's certainly a lot more plausible than in e.g. s6
<kc8apf> that should be doable
<kc8apf> does have some challenging constraints though
indy has quit [Quit: ZNC - http://znc.sourceforge.net]
<awygle> azonenberg: thinking of rack mounting my test equipment. Think I can get away with a two post desk rack or do I need a full four-post?
<azonenberg> I have a 2-post rack for all of my test equipment
<azonenberg> on my desk
<azonenberg> nothing super heavy though
<azonenberg> Down the road i plan to move to two racks because i expect to get more gear
<azonenberg> and one rack high enough would put some stuff out of easy reach
indy has joined ##openfpga
<awygle> Arright cool
<awygle> Two post racks make me nervous so I wanted to double check lol
<azonenberg> mine has a pretty big and solid base plate
<azonenberg> its not going anywhere short of an earthquake
<azonenberg> and once i move to the new location, it's going to be bolted to the table
<azonenberg> Which will be bolted to the concrete floor
<azonenberg> so it would take a pretty substantial quake to knock over :p
<awygle> my apartment won't take kindly to lag bolts in their floors
<awygle> rack mount kits are stupidly expensive for what they are..
<sorear> dissolvable glue? suction cups?
<azonenberg> yes they're expensive
<azonenberg> and thats one of the reasons i bought a place
<azonenberg> If i really wanted to go all the way i'd jackhammer out the concrete
<azonenberg> then cast threaded rods into the floor
<azonenberg> but at that point i might as well re-cast the entire floor
<azonenberg> which is a bit further than i want to go at this point :p
<rqou> azonenberg: have time to peek at the xc9500 stuff linked above?
<azonenberg> Nope, actually about to call my bank
<azonenberg> Somebody's been using "my" CC to buy things in Boston
<azonenberg> they flagged a $900ish nightclub charge as fishy but there were a couple of smaller ones for food and parking that went through
<azonenberg> Wonder how it got pwned
<azonenberg> gas station skimmer and/or POS malware are the obvious guesses
<rqou> typical
<rqou> no matter how upset brian krebs gets, I've basically given up on us banking security
<azonenberg> Yeah, i'm waiting for the day when we have EMV readers on PCs
<azonenberg> or better yet emv-ng or whatever they call its successor
<azonenberg> i want amazon to get a signed blob from the secure element on my card every time i buy something
<azonenberg> or have the charge not clear
<azonenberg> maybe allow you to sign a blanket authorization for recurring charges up to $x or something
<azonenberg> but certainly it should not be possible to make a purchase with only the account number
<azonenberg> the us banking system confuses identification and authorization and this leads to all kinds of problems that PKI was designed to solve
<azonenberg> one account number for checks, ACH deposit, ACH withdrawal
<azonenberg> one account number for any and all credit card purchases
<sorear> i wonder if MITM fraud will be much harder to reverse than current forms of fraud
<azonenberg> any time you give that info to a merchant, they have all the info they need to clone your card
<azonenberg> this is a fundamentally broken system
<sorear> next-gen endpoint malware waits for you to visit amazon, then inserts itself between the browser and the EMV reader so that the card signs a malicious transaction instead of your amazon cart, because trusted displays are apparently still too expensive for a card that $hundreds of fees will be paid over
<azonenberg> Yes but that's a win
<azonenberg> it requires the endpoint to be compromised
<azonenberg> i.e. your pc
<azonenberg> or amazon's server but that's a hard target
<azonenberg> current gen, compromise of any merchant anywhere is a win for the attacker, and lets them siphon up zillions of cards at once
<azonenberg> rather than this, which requires pwning client PCs one at a time
<azonenberg> And i keep things pretty locked down in virtual machines, my actual PC doesnt even access the internet at all
<azonenberg> I'm talking to you guys over VNC to a VM in the DMZ that is only used for irc
<azonenberg> i have another vm for online banking, another for general browsing, another for social media, etc
<azonenberg> If we ever got credit card stuff to the point that it actually required a card present for transactions on a PC, i'd have a dedicated VM just for that and probably revert it every day or two to keep it clean
<azonenberg> My goal was, and continues to be, to be a hard enough target that if my cc/banking info ever got pwned it would be somebody else's fault and not mine
<awygle> I am continually surprised ive never had this problem
<azonenberg> me too, i'm surprised it took so long
<azonenberg> I have a suspicion as to what it was too
<awygle> I take almost no precautions besides jiggling suspicious readers
<azonenberg> a couple weeks ago we started buying bulk building materials from the home depot
<azonenberg> And when we did that, we rented a truck from them
<azonenberg> Which had to be returned full of gas
<azonenberg> So i went to the nearby gas station several times when we were renting the truck
<azonenberg> I think one of the pumps there has a skimmer
<azonenberg> The new generation gas pump skimmers live inside the pump, they're invisible on the outside
<azonenberg> they MITM the signal between the magstripe reader and the modem/brain card
<azonenberg> its not a second reader head like the atm shim skimmers etc
<azonenberg> without taking the pump apart there's no way to know it's there
<azonenberg> Other possibility is malware on a POS system, which does happen but i havent been going to any different retailers than usual
<azonenberg> and this must be a recent leak (not an old one that got dumped ages ago and recently got exploited) because i just got issued a new card like a month or two ago
<azonenberg> b/c the old one expired
<sorear> seems like if I can install mitm hardware in gas pumps, it can redirect 1 in 10 EMV scans to fraudulent transactions
<azonenberg> sorear: yeah i dont know much about how emv authenticates merchants etc
<azonenberg> But the good news is, if you did that
<azonenberg> it would only affect that scan (in theory)
<azonenberg> i.e. you wouldnt have to throw away the card
<azonenberg> just reverse that charge
<sorear> my big question at this point is "are EMV-signed transactions still reversable?"
<azonenberg> I believe any charge can be disputed but emv-signed ones take a bit more work and i think there's some shift of liability to the merchant?
<azonenberg> either way the consumer always has the ability to back out
mumptai has joined ##openfpga
<awygle> i knew someone once who thought using credit cards was dangerous but using debit was fine
<azonenberg> yeah, thats the exact opposite
<azonenberg> i generally compartmentalize things pretty well though
<azonenberg> like, i have a checking account that i only use for wire transfers
<azonenberg> it's kept basically empty
<azonenberg> i move stuff to it then ach it out
<azonenberg> or vice versa
<azonenberg> So if the account is pwned, the damage is massively limited
<awygle> I have one of those for sending money to and from my parents
user10032 has joined ##openfpga
m_t has joined ##openfpga
pie_ has joined ##openfpga
pie___ has quit [Ping timeout: 260 seconds]
mumptai has quit [Quit: Verlassend]
wpwrak has quit [Read error: Connection reset by peer]
wpwrak has joined ##openfpga
RaivisR has joined ##openfpga
RaivisR__ has quit [Read error: Connection reset by peer]
rohitksingh has joined ##openfpga
pie_ has quit [Ping timeout: 264 seconds]
eduardo__ has joined ##openfpga
eduardo_ has quit [Ping timeout: 265 seconds]
nickjohnson has quit [Ping timeout: 256 seconds]
nickjohnson has joined ##openfpga
rohitksingh has quit [Quit: Leaving.]
digshadow has joined ##openfpga
gnufan has quit [Ping timeout: 260 seconds]
pie_ has joined ##openfpga
gnufan has joined ##openfpga
gnufan has quit [Ping timeout: 264 seconds]
gnufan has joined ##openfpga
gnufan has quit [Ping timeout: 264 seconds]
gnufan has joined ##openfpga
ym has joined ##openfpga
pie_ has quit [Ping timeout: 240 seconds]
Wallbraker has quit [Quit: removing from IRC because user idle on matrix for 30+ days]
pie_ has joined ##openfpga
ZipCPU|Laptop has joined ##openfpga
AlexDaniel` has quit [Changing host]
AlexDaniel` has joined ##openfpga
RaivisR_ has joined ##openfpga
RaivisR has quit [Read error: Connection reset by peer]
plaes has quit [Ping timeout: 240 seconds]
plaes has joined ##openfpga
plaes has joined ##openfpga
m_w has joined ##openfpga
pie_ has quit [Read error: Connection reset by peer]
ZipCPU|Laptop has quit [Ping timeout: 256 seconds]
finsternis has quit [Ping timeout: 256 seconds]
finsternis has joined ##openfpga
digshadow has quit [Read error: Connection reset by peer]
digshadow1 has joined ##openfpga
m_w has quit [Quit: leaving]
futarisIRCcloud has joined ##openfpga