Re:Formal CHERI: rigorous engineering and design-time proof ...
by pizlonator on Wednesday August 28, 2024 @02:36PM
(#41382713)
Fil-C is capability based and its capabilities offer stronger guarantees than CHERI’s.
My point is that if a single dude working solo for 10 months gets to 6x overhead in the bad case then the overhead of the software capability approach (either exactly Fil-C or something vaguely like it) is likely to be much much less than 6x in the limit. Fil-C was 200x slower just six months ago and I’m not even done optimizing. That should give you an idea of where it’s heading.
Re:Formal CHERI: rigorous engineering and design-time proof ...
by yosefk on Wednesday August 28, 2024 @02:43PM
(#41382786)
it would be interesting if you shared more details
Re:Formal CHERI: rigorous engineering and design-time proof ...
by pizlonator on Wednesday August 28, 2024 @02:56PM
(#41382928)
https://github.com/pizlonator/llvm-project-deluge/blob/delug...
https://m.youtube.com/watch?v=JRoX9_lXJFg
Note that both that doc and the video have old perf numbers. I’m landing optimizations all the time.
Also CHERI isn't just about memory safety. It also supports software isolation & capabilities.
See CherIoT for an example of both of these.