1. Home
  2. Field-Programmable Gate Arrays
  3. damage Atmel AT40k/AT94k with wrong bitstream?

damage Atmel AT40k/AT94k with wrong bitstream?

Hey, does anybody know if you can damage Atmel's newer FPGAs with a bad bitstream (ie vdd-to-ground contention)?

There's no Big Scary Warning in the data sheet (that I could find), but on the other hand, given the fact that so much of the global routing is based on pass transistors and they let you create multi-driver buses, I can't see how they could possibly protect against this.

This was kind of weird. I know Xilinx has put the Big Scary Warning on datasheets for parts which actually can't be damaged this way (covering their ass?), so I'd expect most vendors to err on the side of caution. Hrm.

Anybody know?

- a

--
"I didn't see it then, but it turned out that getting fired was the
 best thing that could have ever happened to me. The heaviness of
 being successful was replaced by the lightness of being a beginner
 again, less sure about everything. It freed me to enter one of the
 most creative periods of my life."

          -- Steve Jobs, commencement speech at Stanford, June 2005
Reply to
Adam Megacz
Loading thread data ...

Just to clarify: A wrong FPGA bitstream can create massive internal contention, which can damage a part. (I have seen a puff of smoke coming out of a totally misconfigured XC3042 15 years ago...) Xilinx has had CRC protection since XC4000, > 12 years ago. CRC protects against accidental errors, but does not protect against feeding a legitimate bitstream, meant for one part type, into the wrong part type. To protect against this, the Xilinx bitstream also checks the chip ID. We have not heard about any problems ever since. Peter Alfke

Reply to
Peter Alfke

This is not ALWAYS true. It was untrue for the XC6200 sieres:

formatting link

But that post (which I just found) explains *why*, and I know that the Atmel chips don't share the same feature (single source per wire).

- a

Reply to
Adam Megacz

The same question: is it possible to damage a Cyclone with a bad bitstream? Remark: it's a feature, not a bug. SRAM-based FPGA devices are too easy to clone. FPGA manufacturers: why can't you mount two chips inside one package, one containing an FPGA and the second one with configuration flash and add readback fuses, JTAG access fuses etc.? A small amount of EEPROM would be great too...

Best regards Piotr Wyderski

Reply to
Piotr Wyderski

Lattice XP? One die is much better than two.

Cheers, Jon

Reply to
Jon Beniston

Or Actel ProASIC. But what about Altera and Xilinx? :-)

Sure, but I've been told that it is hard to produce flash and FPGA cores on a single piece of silicon because of technological differences.

Best regards Piotr Wyderski

Reply to
Piotr Wyderski

AT94Sxx Secure FPSLIC?

--
A. P. Richelieu
Reply to
Ulf Samuelsson

Sure; I'm working mainly with the AT94xx, which is just an AT94Sxx without the EEPROM. Which, in turn, is just an AT40xx plus the AVR microcontroller. Atmel hasn't really changed their fabric since '98.

- a

--
"I didn't see it then, but it turned out that getting fired was the
 best thing that could have ever happened to me. The heaviness of
 being successful was replaced by the lightness of being a beginner
 again, less sure about everything. It freed me to enter one of the
 most creative periods of my life."

          -- Steve Jobs, commencement speech at Stanford, June 2005
Reply to
Adam Megacz

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.