Running old binaries that abuses setjmp/longjmp with recent glibc

Viktor Ostashevskyi ostash@ostash.kiev.ua
Fri Jan 17 22:05:00 GMT 2020


Hello,

пт, 17 січ. 2020 о 22:52 Florian Weimer <fweimer@redhat.com> пише:
> We could perhaps add a tunable that allows the user to disable the
> hardening at process startup, e.g., instead of XORing with a random
> value, always XOR with zero.  The cost for that would be quite small.

Wasn't it you who removed such tunable in
https://sourceware.org/git/?p=glibc.git;a=commit;h=a014cecd82b71b70a6a843e250e06b541ad524f7
? :)
And XOR'ing with 0 won't help, I've tried by reverting your commit
setting LD_POINTER_GUARD=1.
Current mangling on i386 applies shift with rotation after XORing. So
at least on i386 pointers were still mangled even with
LD_POINTER_GUARD=1.

-- 
З повагою, Осташевський Віктор



More information about the Libc-help mailing list