Collision-resistant hashing

Jim Dehnert dehnert at baalbek.engr.sgi.com
Wed Dec 1 02:23:32 UTC 1999


I'm intrigued by Jonathon's idea of separating the link name from the
"real" mangled name.  Suppose we used the collision-resistant hash to
generate the names in the linker symbol table from the agreed-upon
mangled names, and included the mangled names in a separate table
to provide de-mangling capability.  This table could include the
hashed name mapping, or we could just apply the hash function to
the list if needed.

Note that this wouldn't save object file space relative to just using
a compact mangling scheme, since we still include the names, but it
would mean that the linker could see only short names.  I could imagine
a linking speed benefit.  Any opinions?

Jim

-		Jim Dehnert  x3-4272




More information about the cxx-abi-dev mailing list