linux-kernel - Re: [PATCH RFC] seccomp: Implement syscall isolation based on memory areas

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <50a9e680-6be1-ff50-5c82-1bf54c7484a9@gmail.com>
Date:   Sun, 31 May 2020 15:56:17 +0300
From:   Paul Gofman <gofmanp@...il.com>
To:     Andy Lutomirski <luto@...capital.net>,
        Gabriel Krisman Bertazi <krisman@...labora.com>
Cc:     linux-mm@...ck.org, linux-kernel@...r.kernel.org,
        kernel@...labora.com, Thomas Gleixner <tglx@...utronix.de>,
        Kees Cook <keescook@...omium.org>,
        Will Drewry <wad@...omium.org>,
        "H . Peter Anvin" <hpa@...or.com>
Subject: Re: [PATCH RFC] seccomp: Implement syscall isolation based on memory
 areas

On 5/31/20 03:59, Andy Lutomirski wrote:
>
> I’m suggesting that the kernel learn how to help you, maybe like this:
>
> prctl(PR_SET_SYSCALL_THUNK, target, address_of_unredirected_syscall, 0, 0, 0, 0);
>
> This would be inherited on clone/fork and cleared on execve.
>
If we are talking about explicit specification of syscall addresses to
be trapped by Wine, the problem here is that we don't have any way of
knowing the exact addresses of syscalls to be redirected. We would need
some way to find those syscalls in the highly obfuscated dynamically
generated code, the whole purpose of which is to prevent  disassembling,
debugging and finding things like that in it. What we do know is that if
a syscall is executed from any memory which Wine allocates for Windows
application then it should be treated as Windows syscall and routed to
the Wine's dispatch function. Those code areas can be dynamically
allocated and deallocated.

If we are talking about explicit specification of syscall addresses not
to be trapped, it might be technically possible but at the moment looks
so messy so might be considered not feasible. Wine has a great number of
external dependencies. Most of them depends on some other libraries in
turn. Loading of those libraries goes out of Wine control. Linux
libraries are allowed to issue direct syscalls from their code. I am not
sure we can depend on them not doing it and always calling the same
glibc wrapper.