I think part of the reason why Zydis did comparably well on this is because diffing against other libs is already an important part of our workflow. Our approach is a lot cruder though, just a C diffing tool per lib + /dev/urand + GNU/Parallel.
-
-
-
Fleece may be also worth a look (https://github.com/dyninst/tools/tree/master/fleece …), cf. "Structured Random Differential Testing of Instruction Decoders" (ftp://ftp.cs.wisc.edu/paradyn/papers/JayFleece.pdf).
Kraj razgovora
Novi razgovor -
-
-
whew. (intel xed author)
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
How did you deal with normalizing strings results :https://github.com/aquynh/capstone/pull/1303#issuecomment-459883642 …
-
Hey! Our normalization of disassembly strings is currently pretty meh -- our differential results mainly rely on error reporting + the decoder reporting how many bytes it actually consumed + visual inspection of the results. Constant/register normalization is planned
Kraj razgovora
Novi razgovor -
-
-
We'll definitely look into replacing our current custom diff tools with mishegos -- great work!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.