[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220519073824.GA2578@worktop.programming.kicks-ass.net>
Date: Thu, 19 May 2022 09:38:24 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Song Liu <songliubraving@...com>
Cc: Song Liu <song@...nel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
bpf <bpf@...r.kernel.org>, Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Luis Chamberlain <mcgrof@...nel.org>,
"Torvalds, Linus" <torvalds@...ux-foundation.org>,
"Edgecombe, Rick P" <rick.p.edgecombe@...el.com>,
Kernel Team <Kernel-team@...com>
Subject: Re: [PATCH bpf-next 2/5] x86/alternative: introduce text_poke_set
On Wed, May 18, 2022 at 06:34:18PM +0000, Song Liu wrote:
>
>
> > On May 18, 2022, at 10:09 AM, Peter Zijlstra <peterz@...radead.org> wrote:
> >
> > On Sun, May 15, 2022 at 10:40:48PM -0700, Song Liu wrote:
> >> Introduce a memset like API for text_poke. This will be used to fill the
> >> unused RX memory with illegal instructions.
> >
> > FWIW, you're going to use it to set INT3 (0xCC), that's not an illegal
> > instruction. INTO (0xCE) would be an illegal instruction (in 64bit
> > mode).
>
> Hmm… we have been using INT3 as illegal/invalid/special instructions in
> the JIT. I guess they are equally good for this job?
INT3 is right, it's just not illegal. Terminology is everything :-)
INT3 is the breakpoint instruction, it raises #BP, an illegal
instruction would raise #UD. Different exception vectors and such.
Powered by blists - more mailing lists