Gentoo Archives: gentoo-hardened

From: Mike Frysinger <vapier@g.o>
To: gentoo-hardened@l.g.o
Subject: Re: [gentoo-hardened] SSP + setjmp() = badness?
Date: Sun, 06 Aug 2006 18:37:26
Message-Id: 200608061435.36176.vapier@gentoo.org
In Reply to: Re: [gentoo-hardened] SSP + setjmp() = badness? by "Peter S. Mazinger"
1 On Sunday 06 August 2006 07:17, Peter S. Mazinger wrote:
2 > > > ssp may or may
3 > > > not be a good idea given how new the 4.x series is, but as Mike said,
4 > > > at least there's an eager upstream to fix any issues.
5 >
6 > toolchain.eclass misses the needed support for gcc-4.1 like ssp
7
8 you cant just make statements like this without details ...
9 `gcc -fstack-protector` works just fine with gcc-4.1
10
11 i imagine you're referring to pie patchsets and split specs ?
12
13 > > I think the major problem we are facing here is how to cleanly upgrade
14 > > from 3.x to 4.x. symbol names have changed. And using the stub/aliases
15 > > method Peter used in uClibc svn allows the __guard to be overwritten.
16 >
17 > How can __guard be overwritten, it is even marked with attribute_relro?
18
19 not overriding the symbol, overriding the value
20 -mike

Replies

Subject Author
Re: [gentoo-hardened] SSP + setjmp() = badness? "Peter S. Mazinger" <ps.m@×××.net>
Re: [gentoo-hardened] SSP + setjmp() = badness? "Peter S. Mazinger" <ps.m@×××.net>