All of lore.kernel.org
 help / color / mirror / Atom feed
From: Topi Miettinen <toiwoton@gmail.com>
To: "Zbigniew Jędrzejewski-Szmek" <zbyszek@in.waw.pl>,
	"Andy Lutomirski" <luto@kernel.org>
Cc: linux-hotplug@vger.kernel.org,
	systemd Mailing List <systemd-devel@lists.freedesktop.org>,
	Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>,
	Jethro Beekman <jethro@fortanix.com>,
	Casey Schaufler <casey@schaufler-ca.com>,
	linux-sgx@vger.kernel.org, "Svahn, Kai" <kai.svahn@intel.com>,
	"Schlobohm, Bruce" <bruce.schlobohm@intel.com>,
	Stephen Smalley <sds@tycho.nsa.gov>,
	Haitao Huang <haitao.huang@linux.intel.com>,
	Ben Hutchings <ben@decadent.org.uk>
Subject: Re: Creating executable device nodes in /dev?
Date: Thu, 19 Nov 2020 18:05:08 +0000	[thread overview]
Message-ID: <ff9aaf77-add3-dbeb-d05e-aedd93d41df0@gmail.com> (raw)
In-Reply-To: <20201119163245.GN7348@in.waw.pl>

On 19.11.2020 18.32, Zbigniew Jędrzejewski-Szmek wrote:
> On Thu, Nov 19, 2020 at 08:17:08AM -0800, Andy Lutomirski wrote:
>> Hi udev people-
>>
>> The upcoming Linux SGX driver has a device node /dev/sgx.  User code
>> opens it, does various setup things, mmaps it, and needs to be able to
>> create PROT_EXEC mappings.  This gets quite awkward if /dev is mounted
>> noexec.
>>
>> Can udev arrange to make a device node executable on distros that make
>> /dev noexec?  This could be done by bind-mounting from an exec tmpfs.
>> Alternatively, the kernel could probably learn to ignore noexec on
>> /dev/sgx, but that seems a little bit evil.
> 
> I'd be inclined to simply drop noexec from /dev by default.
> We don't do noexec on either /tmp or /dev/shm (because that causes immediate
> problems with stuff like Java and cffi). And if you have those two at your
> disposal anyway, having noexec on /dev doesn't seem important.

I'd propose to not enable exec globally, but if a service needs SGX, it 
could use something like MountOptions=/dev:exec only in those cases 
where it's needed. That way it's possible to disallow writable and 
executable file systems for most services (which typically don't need 
/tmp or /dev/shm either). Of course the opposite 
(MountOptions=/dev:noexec) would be also possible, but I'd expect that 
this would be needed to be used more often.

-Topi

WARNING: multiple messages have this Message-ID (diff)
From: Topi Miettinen <toiwoton@gmail.com>
To: "Zbigniew Jędrzejewski-Szmek" <zbyszek@in.waw.pl>,
	"Andy Lutomirski" <luto@kernel.org>
Cc: linux-hotplug@vger.kernel.org,
	systemd Mailing List <systemd-devel@lists.freedesktop.org>,
	Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>,
	Jethro Beekman <jethro@fortanix.com>,
	Casey Schaufler <casey@schaufler-ca.com>,
	linux-sgx@vger.kernel.org, "Svahn, Kai" <kai.svahn@intel.com>,
	"Schlobohm, Bruce" <bruce.schlobohm@intel.com>,
	Stephen Smalley <sds@tycho.nsa.gov>,
	Haitao Huang <haitao.huang@linux.intel.com>,
	Ben Hutchings <ben@decadent.org.uk>
Subject: Re: Creating executable device nodes in /dev?
Date: Thu, 19 Nov 2020 20:05:08 +0200	[thread overview]
Message-ID: <ff9aaf77-add3-dbeb-d05e-aedd93d41df0@gmail.com> (raw)
In-Reply-To: <20201119163245.GN7348@in.waw.pl>

On 19.11.2020 18.32, Zbigniew Jędrzejewski-Szmek wrote:
> On Thu, Nov 19, 2020 at 08:17:08AM -0800, Andy Lutomirski wrote:
>> Hi udev people-
>>
>> The upcoming Linux SGX driver has a device node /dev/sgx.  User code
>> opens it, does various setup things, mmaps it, and needs to be able to
>> create PROT_EXEC mappings.  This gets quite awkward if /dev is mounted
>> noexec.
>>
>> Can udev arrange to make a device node executable on distros that make
>> /dev noexec?  This could be done by bind-mounting from an exec tmpfs.
>> Alternatively, the kernel could probably learn to ignore noexec on
>> /dev/sgx, but that seems a little bit evil.
> 
> I'd be inclined to simply drop noexec from /dev by default.
> We don't do noexec on either /tmp or /dev/shm (because that causes immediate
> problems with stuff like Java and cffi). And if you have those two at your
> disposal anyway, having noexec on /dev doesn't seem important.

I'd propose to not enable exec globally, but if a service needs SGX, it 
could use something like MountOptions=/dev:exec only in those cases 
where it's needed. That way it's possible to disallow writable and 
executable file systems for most services (which typically don't need 
/tmp or /dev/shm either). Of course the opposite 
(MountOptions=/dev:noexec) would be also possible, but I'd expect that 
this would be needed to be used more often.

-Topi

  reply	other threads:[~2020-11-19 18:05 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-19 16:17 Creating executable device nodes in /dev? Andy Lutomirski
2020-11-19 16:17 ` Andy Lutomirski
2020-11-19 16:32 ` Zbigniew Jędrzejewski-Szmek
2020-11-19 18:05   ` Topi Miettinen [this message]
2020-11-19 18:05     ` Topi Miettinen
2020-12-08 18:07     ` Andy Lutomirski
2020-12-08 18:07       ` Andy Lutomirski
2020-12-08 20:45       ` Topi Miettinen
2020-12-08 20:45         ` Topi Miettinen
2020-12-08 21:30         ` Andy Lutomirski
2020-12-08 21:30           ` Andy Lutomirski
2020-12-08 23:15           ` Topi Miettinen
2020-12-08 23:15             ` Topi Miettinen
2020-12-09  0:15             ` Jarkko Sakkinen
2020-12-09  0:15               ` Jarkko Sakkinen
2020-12-09  0:42               ` Jarkko Sakkinen
2020-12-09  0:42                 ` Jarkko Sakkinen
2020-12-09  8:58                 ` Topi Miettinen
2020-12-09  8:58                   ` Topi Miettinen
2020-12-09  9:07                   ` Jethro Beekman
2020-12-09  9:07                     ` Jethro Beekman
2020-12-09 15:14                   ` Andy Lutomirski
2020-12-09 15:14                     ` Andy Lutomirski
2020-12-09 19:22                     ` Topi Miettinen
2020-12-09 19:22                       ` Topi Miettinen
2020-12-09 19:32                       ` Andy Lutomirski
2020-12-09 19:32                         ` Andy Lutomirski
2020-12-09 21:58                     ` Ben Hutchings
2020-12-09 21:58                       ` Ben Hutchings
2020-12-11 11:36                       ` Zbigniew Jędrzejewski-Szmek
2020-12-09  7:58               ` Antw: [EXT] Re: [systemd-devel] " Ulrich Windl
2020-12-09  7:58                 ` Ulrich Windl
2020-12-11 10:40                 ` Jarkko Sakkinen
2020-12-11 10:40                   ` Jarkko Sakkinen
2020-12-09  8:35               ` Topi Miettinen
2020-12-09  8:35                 ` Topi Miettinen
2020-12-11 10:46                 ` Jarkko Sakkinen
2020-12-11 10:46                   ` Jarkko Sakkinen
2020-12-11 11:29                   ` Greg KH
2020-12-11 11:29                     ` Greg KH
2020-12-12 11:51                     ` [systemd-devel] " Christian Brauner
2020-12-12 11:51                       ` Christian Brauner
2020-12-12 12:32                     ` Christian Brauner
2020-12-12 12:32                       ` Christian Brauner
2020-12-11 11:46                   ` Topi Miettinen
2020-12-11 11:46                     ` Topi Miettinen
2020-12-14  7:25                     ` Antw: [EXT] Re: [systemd-devel] " Ulrich Windl
2020-12-14  7:25                       ` Ulrich Windl
2020-12-15  4:19                       ` Jarkko Sakkinen
2020-12-15  4:19                         ` Jarkko Sakkinen
2020-12-15  4:27                         ` Jarkko Sakkinen
2020-12-15  4:27                           ` Jarkko Sakkinen
2020-12-16 10:03                         ` Ulrich Windl
2020-12-16 10:03                           ` Ulrich Windl
2020-12-16 13:05                           ` Topi Miettinen
2020-12-16 13:05                             ` Topi Miettinen
2020-12-22 22:14                             ` Jarkko Sakkinen
2020-12-22 22:14                               ` Jarkko Sakkinen
2020-12-09  0:03       ` Jarkko Sakkinen
2020-12-09  0:03         ` Jarkko Sakkinen

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 \
    --in-reply-to=ff9aaf77-add3-dbeb-d05e-aedd93d41df0@gmail.com \
    --to=toiwoton@gmail.com \
    --cc=ben@decadent.org.uk \
    --cc=bruce.schlobohm@intel.com \
    --cc=casey@schaufler-ca.com \
    --cc=haitao.huang@linux.intel.com \
    --cc=jarkko.sakkinen@linux.intel.com \
    --cc=jethro@fortanix.com \
    --cc=kai.svahn@intel.com \
    --cc=linux-hotplug@vger.kernel.org \
    --cc=linux-sgx@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=sds@tycho.nsa.gov \
    --cc=systemd-devel@lists.freedesktop.org \
    --cc=zbyszek@in.waw.pl \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.