[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5772EE62.11219.7B892887@pageexec.freemail.hu>
Date: Tue, 28 Jun 2016 23:38:42 +0200
From: "PaX Team" <pageexec@...email.hu>
To: Emese Revfy <re.emese@...il.com>,
Rasmus Villemoes <linux@...musvillemoes.dk>
CC: kernel-hardening@...ts.openwall.com, spender@...ecurity.net,
mmarek@...e.com, keescook@...omium.org,
linux-kernel@...r.kernel.org, yamada.masahiro@...ionext.com,
linux-kbuild@...r.kernel.org, minipli@...linux.so,
linux@...linux.org.uk, catalin.marinas@....com,
david.brown@...aro.org, benh@...nel.crashing.org,
tglx@...utronix.de, akpm@...ux-foundation.org,
jlayton@...chiereds.net, arnd@...db.de
Subject: Re: [PATCH v1 2/2] Mark functions with the __nocapture attribute
On 28 Jun 2016 at 22:50, Rasmus Villemoes wrote:
> > +extern const char *kstrdup_const(const char *s, gfp_t gfp) __nocapture(1);
>
> OK, so this one is pretty dangerous, and probably wrong. If one does
>
> foo->bar = kstrdup_const(a-macro-that-might-be-a-string-literal)
>
> in an .init function, foo->bar will very likely become dangling.
doesn't kstrdup_const omit the copy only for arguments that are stored in
.rodata (which doesn't include .init.rodata* and other init sections)?
cheers,
PaX Team
Powered by blists - more mailing lists