https://tomverbeure.github.io/2019/11/21/Z-tape.html
(A bit off topic but that Bloomberg story is still such a mystery to me. It went into quite a bit of details and claimed it had tons of sources, so even if it was a completely false story whoever fabricated it must have put insane amounts of efforts/identity fraud into the set up)
Could the device not simply get the host to do these things for it, by e.g. rootkit-ing the server’s BMC? A “hardware virus”, per se.
> Could the device not simply get the host to do these things for it, by e.g. rootkit-ing the server’s BMC? A “hardware virus”, per se.
IIRC, that's exactly what was alleged in the article. It was an implant that sat on the BMC ROM bus, fiddling with bits as the ROM was read during bootup. No need for any networking or processing ability beyond what was needed to that. This guy actually did a POC of that: https://trmm.net/Modchips/.
So totally plausable.
That would require some processing chops to handle whatever speed the SPI bus ran at, and a bit of space to store the replacement bytes. Firmly within the margin for plausibility with even basic off-the-shelf kit. Honestly depressing really.
Hold them on with one hand while you type a few commands over the serial connection with the other.
If you have more in depth debugging to do, get a 'helping hand' to hold it in place.
(The UART provided a root shell and enough power to “boot” the ATtiny which simply waited a few seconds and then ran some commands to initiate a reverse shell to a server under our control every time the device was powered on. Thanks to this tape and the device’s tool-less case (and convenient unpopulated header with space around it), it was enlightening how trivially easy it’d be to develop and deploy such an implant to an operating device (with the caveat that I wouldn’t consider the connection robust enough to survive transport).
It’s also useful to connect SMD EEPROMs to unpopulated/desoldered pads for testing without installing a socket.