[Cryptech Tech] Alpha board block diagram v0.001

Joachim Strömbergson joachim at secworks.se
Tue Mar 24 10:34:46 UTC 2015


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Aloha!

Павел Шатов wrote:
> - Give an estimate of total power consumption. How much current @
> 12V will the board require?

Good point. I've added a note about that in the component document.


> Turns out, we will not be able to have separate address and 32-bit
> data bus. The higher part of data bus (signals EIM_D[16]..EIM_D[31])
> are multiplexed with SPI and UART modules in iMX6Q. We can either
> have SPI and UART, or 32-bit EIM. In this situation I suggest using
> separate 24-bit address bus and 16-bit address bus. At 66 MHz this
> will give us 4 times more bandwidth, than we have now.

Dang. I assume you mean 24-bit address and 16-bit data. As you say, it
is still better than a single bus for address and data and double the
frequency. We take what we can. I'll update the text and diagram.

> 
> 3. USB-to-Serial Converter
> 
> I'm afraid, MAX3421E is not what we need. It is not a transceiver, it
> is a Peripheral/Host controller. It does have SPI interface, but
> this interface is used to communicate with its internal registers,
> not to communicate with external system. I believe, what we actually
> need is FT232R from FTDI or maybe CP2104 from SiLabs.

I've seen the following discussions re UART vs SPI. The big difference
in general is the capacity you get. SPI being source clocked allows you
to have 10-15 Mbps with SPI while the FTDI USB-UART chip tops out at 3
Mbps. And the i.mx6 UARTS tops out at 4 Mbps.

The problem with SPI is to find a simple transciever-like component. The
MAX3421E _can_ actually drive external SPI. We have used it in a project
to do just that. But it is a complex chip. We are now using the FDTI
FT232H which is simpler and cheaper:

http://www.ftdichip.com/Support/Documents/DataSheets/ICs/DS_FT232H.pdf

We get 12 Mbps USB-SPI performance to our SPI slave (sensor and MCU)
from SW running on the host using the FTDI virtual serial ports. But
this chip is still not just a transciever.

This is where we should look at the use cases and what host
communication performance we think we need for the alpha board.
Randy and Rob, what do you think?

If we think max 3 Mbps is fine then lets use the FT232R suggested by
Pavel. But if we think we need more we might need to rethink our host
interface. We can:

(1) Use the USB interface in the i.mx6 directly

(2) Use something like the FT232H. More complex as it is and thus less
trustworthy, but provides faster access.

(3) Look again at some other interface. Rob suggested raw Ethernet and
that could be achieved in different ways: External PHY and simple 10/100
Mbps MAC in the FPGA, external MAC+PHY as either combines or separate
components. Or use the Ethernet-IF in the i.mx6.


For the management interface, using USB-UART should be ok I believe.
- -- 
Med vänlig hälsning, Yours

Joachim Strömbergson - Alltid i harmonisk svängning.
========================================================================
 Joachim Strömbergson          Secworks AB          joachim at secworks.se
========================================================================
-----BEGIN PGP SIGNATURE-----
Version: GnuPG/MacGPG2 v2
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQIcBAEBCAAGBQJVET3GAAoJEF3cfFQkIuyNL6MP/2cpZDCIngHsb988/w68EHTz
jTTzRgaQNArezNq0zxBwIA51cPd8ATswp/upV3JSPmFDD0a6XcxR0Fd7Nri2uba3
sjCFX8bEjcqe21h4yNQl8WRrue5mZPDWp0zuJ2P4MmqRq5irgDqN88oaa71coEJz
zJ/8NVOOx2gLbJFg+t0KpphOfG/2Ppwv9SJCMz3zxiWm+GpC4s18c83KWvh3x69v
lWz1NRwSBoB8GNfL1Hq3jinyz18ZJNQCNOdrMq6DzmU/thwuz8zZ0KHQHovfAyvP
ZwkBZAMzN8IqzWO6jC9pxnHTdPnjePk1vuRxYV8G1zqCWYoK/Ba0cgabmneyBEmk
HFKdTdaeMFhY9O9PQ4skT/opB5LSH5mA451BOH/49y1Qr7Ycs84xmxZZMmZWZOsk
kTiEZCPvYz0N0wFK4bsyecYjZVDtaEsK3JGiTSbqqfOmGEZUJ6B8OdXpPIrW7UaE
YUA1ZuZi08TGbwcKjm5U9hLMo9iwyLCMCOoShdkVcKp75F+nW1LU39cr5V7sb9Jl
ViLt+Wr+NydZqENQ3+dZYy4dZTCxerEZaPvLh6He5O45MQ4vfye6sBixvPf2SflA
lZAbz2Kl7R1qIWIMBZCg18caReFO0ArILTAD3R6B+M5d6MeSWVH2kSKmhJRsWc59
KMIWd1Y2sfnq8oLhr4t1
=yODq
-----END PGP SIGNATURE-----


More information about the Tech mailing list