Re: [PATCH] debugfs: fix security_locked_down() call for SELinux

[Matthew Wilcox]

On Fri, May 07, 2021 at 02:03:04PM +0200, Greg Kroah-Hartman wrote:
> On Fri, May 07, 2021 at 01:41:50PM +0200, Ondrej Mosnacek wrote:
> > Make sure that security_locked_down() is checked last so that a bogus
> > denial is not reported by SELinux when (ia->ia_valid & (ATTR_MODE |
> > ATTR_UID | ATTR_GID)) is zero.
> 
> Why would this be "bogus"?

I presume selinux is logging a denial ... but we don't then actually
deny the operation.

> > Note: this was introduced by commit 5496197f9b08 ("debugfs: Restrict
> > debugfs when the kernel is locked down"), but it didn't matter at that
> > time, as the SELinux support came in later.
> > 
> > Fixes: 59438b46471a ("security,lockdown,selinux: implement SELinux lockdown")
> 
> What does this "fix"?
> 
> What is happening in selinux that it can not handle this sequence now?
> That commit showed up a long time ago, this feels "odd"...
> 
> thanks,
> 
> greg k-h