Re: [PATCH -v1 2/3] vm: use new has_capability_noaudit

[Eric Paris <eparis@redhat.com>]

On Wed, 2008-10-29 at 15:15 -0400, Stephen Smalley wrote:
> On Wed, 2008-10-29 at 15:06 -0400, Eric Paris wrote:
> > The oomkiller calculations make decisions based on capabilities.  Since
> > these are not security decisions and LSMs should not record if they fall
> > the request they should use the new has_capability_noaudit() interface so
> > the denials will not be recorded.
> > 
> > Signed-off-by: Eric Paris <eparis@redhat.com>
> > ---
> > 
> >  fs/proc/base.c |    2 +-
> >  mm/oom_kill.c  |    6 +++---
> >  2 files changed, 4 insertions(+), 4 deletions(-)
> > 
> > diff --git a/fs/proc/base.c b/fs/proc/base.c
> > index 486cf3f..ef83e81 100644
> > --- a/fs/proc/base.c
> > +++ b/fs/proc/base.c
> > @@ -1020,7 +1020,7 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf,
> >  	task = get_proc_task(file->f_path.dentry->d_inode);
> >  	if (!task)
> >  		return -ESRCH;
> > -	if (oom_adjust < task->oomkilladj && !capable(CAP_SYS_RESOURCE)) {
> > +	if (oom_adjust < task->oomkilladj && !has_capability_noaudit(current, CAP_SYS_RESOURCE)) {
> 
> This one looks like an actual permission check to see whether the
> current task is authorized to modify this value (by writing to some proc
> node).  Which should be audited.  Unlike the others, where they are
> checking whether some other task has a capability in order to help
> decide priorities for the OOM killer.

Will be fixed in -v2