A mysterious bug in the firmware of Google's Titan M chip

Re:A mysterious bug in the firmware of Google's Titan M chip by _Microft on Saturday February 29, 2020 @01:39PM (#22452121)

Could it be that the first excerpt (https://github.com/beemdevelopment/Aegis/blob/def7c676148f26... ) should not have included the line with .setIsStrongBoxBacked(true) yet?

Edit: funny: 0dd0adde looks a lot like deadd00d (like "dead dude") with a reversed byte order. If this is a coincidence?

Edit2: https://www.bing.com/search?q=deadd00d

[ Parent | on HN ]

Re:A mysterious bug in the firmware of Google's Titan M chip by v64 on Saturday February 29, 2020 @02:12PM (#22452352)

Very interesting observation!

"Comments on a number of StackOverflow questions have pointed out that a fault address of deadd00d indicates a deliberate VM abort."

That ending up in the ciphertext multiple times seems to point to some memory corruption issue. It's also a good argument for using magic numbers that stand out.

[ Parent | on HN ]

Re:A mysterious bug in the firmware of Google's Titan M chip by alexbakker on Saturday February 29, 2020 @01:42PM (#22452135)

You're absolutely right about the excerpt. Fixed, thanks.

[ Parent | on HN ]