mbox series

[v5.4,v2,0/4] Update SELinuxfs out of tree and then swapover

Message ID 20201016134835.1886478-1-dburgener@linux.microsoft.com (mailing list archive)
Headers show
Series Update SELinuxfs out of tree and then swapover | expand

Message

Daniel Burgener Oct. 16, 2020, 1:48 p.m. UTC
v2: Include all commits from original series, and include commit ids

This is a backport for stable of my series to fix a race condition in
selinuxfs during policy load:

selinux: Create function for selinuxfs directory cleanup
https://lore.kernel.org/selinux/20200819195935.1720168-2-dburgener@linux.microsoft.com/

selinux: Refactor selinuxfs directory populating functions
https://lore.kernel.org/selinux/20200819195935.1720168-3-dburgener@linux.microsoft.com/

selinux: Standardize string literal usage for selinuxfs directory names
https://lore.kernel.org/selinux/20200819195935.1720168-4-dburgener@linux.microsoft.com/

selinux: Create new booleans and class dirs out of tree
https://lore.kernel.org/selinux/20200819195935.1720168-5-dburgener@linux.microsoft.com/

Several changes were necessary to backport.  They are detailed in the
commit message for the third commit in the series.

The bulk of the original cover letter is reproduced below.

In the current implementation, on policy load /sys/fs/selinux is updated
by deleting the previous contents of
/sys/fs/selinux/{class,booleans} and then recreating them.  This means
that there is a period of time when the contents of these directories do
not exist which can cause race conditions as userspace relies on them for
information about the policy.  In addition, it means that error recovery
in the event of failure is challenging.

This patch series follows the design outlined by Al Viro in a previous
e-mail to the list[1].  This approach is to first create the new
directory structures out of tree, then to perform the swapover, and
finally to delete the old directories.  Not handled in this series is
error recovery in the event of failure.

Error recovery in the selinuxfs recreation is unhandled in the current
code, so this series will not cause any regression in this regard.
Handling directory recreation in this manner is a prerequisite to make
proper error handling possible.

In order to demonstrate the race condition that this series fixes, you
can use the following commands:

while true; do cat /sys/fs/selinux/class/service/perms/status
>/dev/null; done &
while true; do load_policy; done;

In the existing code, this will display errors fairly often as the class
lookup fails.  (In normal operation from systemd, this would result in a
permission check which would be allowed or denied based on policy settings
around unknown object classes.) After applying this patch series you
should expect to no longer see such error messages.

Daniel Burgener (4):
  selinux: Create function for selinuxfs directory cleanup
  selinux: Refactor selinuxfs directory populating functions
  selinux: Standardize string literal usage for selinuxfs directory
    names
  selinux: Create new booleans and class dirs out of tree

 security/selinux/selinuxfs.c | 168 +++++++++++++++++++++++++++--------
 1 file changed, 129 insertions(+), 39 deletions(-)

Comments

Greg Kroah-Hartman Oct. 16, 2020, 3:01 p.m. UTC | #1
On Fri, Oct 16, 2020 at 09:48:31AM -0400, Daniel Burgener wrote:
> v2: Include all commits from original series, and include commit ids
> 
> This is a backport for stable of my series to fix a race condition in
> selinuxfs during policy load:

Has this race condition always been present, or is this a regression
that is being fixed from previously working kernels?

If it's always been present, why not just use 5.9 to solve it?

thanks,

greg k-h
Sasha Levin Oct. 16, 2020, 3:38 p.m. UTC | #2
On Fri, Oct 16, 2020 at 05:01:20PM +0200, Greg KH wrote:
>On Fri, Oct 16, 2020 at 09:48:31AM -0400, Daniel Burgener wrote:
>> v2: Include all commits from original series, and include commit ids
>>
>> This is a backport for stable of my series to fix a race condition in
>> selinuxfs during policy load:
>
>Has this race condition always been present, or is this a regression
>that is being fixed from previously working kernels?

So this issue has always been there, but:

>If it's always been present, why not just use 5.9 to solve it?

Because it was merged for 5.10 rather than 5.9, which is a few months
out, so Daniel is looking to see if we can have it in 5.8/5.4 to close
the gap.
Greg Kroah-Hartman Oct. 16, 2020, 3:44 p.m. UTC | #3
On Fri, Oct 16, 2020 at 11:38:23AM -0400, Sasha Levin wrote:
> On Fri, Oct 16, 2020 at 05:01:20PM +0200, Greg KH wrote:
> > On Fri, Oct 16, 2020 at 09:48:31AM -0400, Daniel Burgener wrote:
> > > v2: Include all commits from original series, and include commit ids
> > > 
> > > This is a backport for stable of my series to fix a race condition in
> > > selinuxfs during policy load:
> > 
> > Has this race condition always been present, or is this a regression
> > that is being fixed from previously working kernels?
> 
> So this issue has always been there, but:
> 
> > If it's always been present, why not just use 5.9 to solve it?
> 
> Because it was merged for 5.10 rather than 5.9, which is a few months
> out, so Daniel is looking to see if we can have it in 5.8/5.4 to close
> the gap.

I would have to wait for 5.10-rc1 at the earliest, and get the selinux
maintainers ack for this :)

thanks,

greg k-h
Sasha Levin Oct. 16, 2020, 3:49 p.m. UTC | #4
On Fri, Oct 16, 2020 at 05:44:43PM +0200, Greg KH wrote:
>On Fri, Oct 16, 2020 at 11:38:23AM -0400, Sasha Levin wrote:
>> On Fri, Oct 16, 2020 at 05:01:20PM +0200, Greg KH wrote:
>> > On Fri, Oct 16, 2020 at 09:48:31AM -0400, Daniel Burgener wrote:
>> > > v2: Include all commits from original series, and include commit ids
>> > >
>> > > This is a backport for stable of my series to fix a race condition in
>> > > selinuxfs during policy load:
>> >
>> > Has this race condition always been present, or is this a regression
>> > that is being fixed from previously working kernels?
>>
>> So this issue has always been there, but:
>>
>> > If it's always been present, why not just use 5.9 to solve it?
>>
>> Because it was merged for 5.10 rather than 5.9, which is a few months
>> out, so Daniel is looking to see if we can have it in 5.8/5.4 to close
>> the gap.
>
>I would have to wait for 5.10-rc1 at the earliest, and get the selinux
>maintainers ack for this :)

No objections; if the selinux folks feel unhappy with this it'll just
wait for 5.10.
Daniel Burgener Oct. 16, 2020, 4:01 p.m. UTC | #5
On 10/16/20 11:49 AM, Sasha Levin wrote:
> On Fri, Oct 16, 2020 at 05:44:43PM +0200, Greg KH wrote:
>> On Fri, Oct 16, 2020 at 11:38:23AM -0400, Sasha Levin wrote:
>>> On Fri, Oct 16, 2020 at 05:01:20PM +0200, Greg KH wrote:
>>> > On Fri, Oct 16, 2020 at 09:48:31AM -0400, Daniel Burgener wrote:
>>> > > v2: Include all commits from original series, and include commit 
>>> ids
>>> > >
>>> > > This is a backport for stable of my series to fix a race 
>>> condition in
>>> > > selinuxfs during policy load:
>>> >
>>> > Has this race condition always been present, or is this a regression
>>> > that is being fixed from previously working kernels?
>>>
>>> So this issue has always been there, but:
>>>
>>> > If it's always been present, why not just use 5.9 to solve it?
>>>
>>> Because it was merged for 5.10 rather than 5.9, which is a few months
>>> out, so Daniel is looking to see if we can have it in 5.8/5.4 to close
>>> the gap.
>>
>> I would have to wait for 5.10-rc1 at the earliest, and get the selinux
>> maintainers ack for this :)
>
> No objections; if the selinux folks feel unhappy with this it'll just
> wait for 5.10.
>
Yes, that's fine from my end as well.  We can carry this series out of 
tree in the interim.  Thanks!

-Daniel
Paul Moore Oct. 16, 2020, 10:56 p.m. UTC | #6
On Fri, Oct 16, 2020 at 12:01 PM Daniel Burgener
<dburgener@linux.microsoft.com> wrote:
> On 10/16/20 11:49 AM, Sasha Levin wrote:
> > On Fri, Oct 16, 2020 at 05:44:43PM +0200, Greg KH wrote:
> >> On Fri, Oct 16, 2020 at 11:38:23AM -0400, Sasha Levin wrote:
> >>> On Fri, Oct 16, 2020 at 05:01:20PM +0200, Greg KH wrote:
> >>> > On Fri, Oct 16, 2020 at 09:48:31AM -0400, Daniel Burgener wrote:
> >>> > > v2: Include all commits from original series, and include commit
> >>> ids
> >>> > >
> >>> > > This is a backport for stable of my series to fix a race
> >>> condition in
> >>> > > selinuxfs during policy load:
> >>> >
> >>> > Has this race condition always been present, or is this a regression
> >>> > that is being fixed from previously working kernels?
> >>>
> >>> So this issue has always been there, but:
> >>>
> >>> > If it's always been present, why not just use 5.9 to solve it?
> >>>
> >>> Because it was merged for 5.10 rather than 5.9, which is a few months
> >>> out, so Daniel is looking to see if we can have it in 5.8/5.4 to close
> >>> the gap.
> >>
> >> I would have to wait for 5.10-rc1 at the earliest, and get the selinux
> >> maintainers ack for this :)
> >
> > No objections; if the selinux folks feel unhappy with this it'll just
> > wait for 5.10.
> >
> Yes, that's fine from my end as well.  We can carry this series out of
> tree in the interim.  Thanks!

I tend to be pretty conservative when it comes to backporting patches
to -stable, and since this is both a) big and b) fixes a problem that
has existed since the dawn of selinuxfs (and possibly longer <g>) I
think the smart thing to do is to wait for v5.10.