[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a599d509-8995-0eb6-65a1-5d8505e4851e@citrix.com>
Date: Wed, 16 Feb 2022 00:57:47 +0000
From: Andrew Cooper <Andrew.Cooper3@...rix.com>
To: Peter Zijlstra <peterz@...radead.org>,
Joao Moreira <joao@...rdrivepizza.com>
CC: Kees Cook <keescook@...omium.org>,
"x86@...nel.org" <x86@...nel.org>,
"hjl.tools@...il.com" <hjl.tools@...il.com>,
"jpoimboe@...hat.com" <jpoimboe@...hat.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"ndesaulniers@...gle.com" <ndesaulniers@...gle.com>,
"samitolvanen@...gle.com" <samitolvanen@...gle.com>,
"llvm@...ts.linux.dev" <llvm@...ts.linux.dev>
Subject: Re: [RFC][PATCH 6/6] objtool: Add IBT validation / fixups
On 11/02/2022 13:38, Peter Zijlstra wrote:
> Now, this setup provides:
>
> - minimal size (please yell if there's a smaller option I missed;
> s/ud2/int3/ ?)
It's probably best not to overload int3. #BP is already has magic
properties, and what happens if someone wants to breakpoint one of these?
I'll make my usual into suggestion for 64bit builds.
ud2 is the sensible approach, with a caveat that whatever is hiding here
should have metadata.
~Andrew
Powered by blists - more mailing lists