From: Stefan Hajnoczi <[email protected]>
To: Stefano Garzarella <[email protected]>
Cc: Andy Lutomirski <[email protected]>, Jens Axboe <[email protected]>,
Christoph Hellwig <[email protected]>, Kees Cook <[email protected]>,
Pavel Begunkov <[email protected]>,
Miklos Szeredi <[email protected]>,
Matthew Wilcox <[email protected]>,
Jann Horn <[email protected]>,
Christian Brauner <[email protected]>,
[email protected], [email protected],
Linux API <[email protected]>,
Linux FS Devel <[email protected]>,
LKML <[email protected]>,
Michael Kerrisk <[email protected]>
Subject: Re: strace of io_uring events?
Date: Thu, 23 Jul 2020 11:39:49 +0100 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <20200721155848.32xtze5ntvcmjv63@steredhat>
[-- Attachment #1: Type: text/plain, Size: 3621 bytes --]
On Tue, Jul 21, 2020 at 05:58:48PM +0200, Stefano Garzarella wrote:
> On Tue, Jul 21, 2020 at 08:27:34AM -0700, Andy Lutomirski wrote:
> > On Fri, Jul 17, 2020 at 1:02 AM Stefano Garzarella <[email protected]> wrote:
> > >
> > > On Thu, Jul 16, 2020 at 08:12:35AM -0700, Kees Cook wrote:
> > > > On Thu, Jul 16, 2020 at 03:14:04PM +0200, Stefano Garzarella wrote:
> >
> > > > access (IIUC) is possible without actually calling any of the io_uring
> > > > syscalls. Is that correct? A process would receive an fd (via SCM_RIGHTS,
> > > > pidfd_getfd, or soon seccomp addfd), and then call mmap() on it to gain
> > > > access to the SQ and CQ, and off it goes? (The only glitch I see is
> > > > waking up the worker thread?)
> > >
> > > It is true only if the io_uring istance is created with SQPOLL flag (not the
> > > default behaviour and it requires CAP_SYS_ADMIN). In this case the
> > > kthread is created and you can also set an higher idle time for it, so
> > > also the waking up syscall can be avoided.
> >
> > I stared at the io_uring code for a while, and I'm wondering if we're
> > approaching this the wrong way. It seems to me that most of the
> > complications here come from the fact that io_uring SQEs don't clearly
> > belong to any particular security principle. (We have struct creds,
> > but we don't really have a task or mm.) But I'm also not convinced
> > that io_uring actually supports cross-mm submission except by accident
> > -- as it stands, unless a user is very careful to only submit SQEs
> > that don't use user pointers, the results will be unpredictable.
> > Perhaps we can get away with this:
> >
> > diff --git a/fs/io_uring.c b/fs/io_uring.c
> > index 74bc4a04befa..92266f869174 100644
> > --- a/fs/io_uring.c
> > +++ b/fs/io_uring.c
> > @@ -7660,6 +7660,20 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int,
> > fd, u32, to_submit,
> > if (!percpu_ref_tryget(&ctx->refs))
> > goto out_fput;
> >
> > + if (unlikely(current->mm != ctx->sqo_mm)) {
> > + /*
> > + * The mm used to process SQEs will be current->mm or
> > + * ctx->sqo_mm depending on which submission path is used.
> > + * It's also unclear who is responsible for an SQE submitted
> > + * out-of-process from a security and auditing perspective.
> > + *
> > + * Until a real usecase emerges and there are clear semantics
> > + * for out-of-process submission, disallow it.
> > + */
> > + ret = -EACCES;
> > + goto out;
> > + }
> > +
> > /*
> > * For SQ polling, the thread will do all submissions and completions.
> > * Just return the requested submit count, and wake the thread if
> >
> > If we can do that, then we could bind seccomp-like io_uring filters to
> > an mm, and we get obvious semantics that ought to cover most of the
> > bases.
> >
> > Jens, Christoph?
> >
> > Stefano, what's your intended usecase for your restriction patchset?
> >
>
> Hi Andy,
> my use case concerns virtualization. The idea, that I described in the
> proposal of io-uring restrictions [1], is to share io_uring CQ and SQ queues
> with a guest VM for block operations.
>
> In the PoC that I realized, there is a block device driver in the guest that
> uses io_uring queues coming from the host to submit block requests.
>
> Since the guest is not trusted, we need restrictions to allow only
> a subset of syscalls on a subset of file descriptors and memory.
BTW there's only a single mm in the kvm.ko use case.
Stefan
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next prev parent reply other threads:[~2020-07-23 10:40 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-15 11:12 strace of io_uring events? Miklos Szeredi
2020-07-15 14:35 ` Andy Lutomirski
2020-07-15 17:11 ` Matthew Wilcox
2020-07-15 19:42 ` Pavel Begunkov
2020-07-15 20:09 ` Miklos Szeredi
2020-07-15 20:20 ` Pavel Begunkov
2020-07-15 23:07 ` Kees Cook
2020-07-16 13:14 ` Stefano Garzarella
2020-07-16 15:12 ` Kees Cook
2020-07-17 8:01 ` Stefano Garzarella
2020-07-21 15:27 ` Andy Lutomirski
2020-07-21 15:31 ` Jens Axboe
2020-07-21 17:23 ` Andy Lutomirski
2020-07-21 17:30 ` Jens Axboe
2020-07-21 17:44 ` Andy Lutomirski
2020-07-21 18:39 ` Jens Axboe
2020-07-21 19:44 ` Andy Lutomirski
2020-07-21 19:48 ` Jens Axboe
2020-07-21 19:56 ` Andres Freund
2020-07-21 19:37 ` Andres Freund
2020-07-21 15:58 ` Stefano Garzarella
2020-07-23 10:39 ` Stefan Hajnoczi [this message]
2020-07-23 13:37 ` Colin Walters
2020-07-24 7:25 ` Stefano Garzarella
2020-07-16 13:17 ` Aleksa Sarai
2020-07-16 15:19 ` Kees Cook
2020-07-17 8:17 ` Cyril Hrubis
2020-07-16 16:24 ` Andy Lutomirski
2020-07-16 0:12 ` tytso
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox