[PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Richard Guy Briggs]

Tie syscall information to FEATURE_CHANGE calls since it is a result of
user action.

See: https://github.com/linux-audit/audit-kernel/issues/80

Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
---
 kernel/audit.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/kernel/audit.c b/kernel/audit.c
index 8da24ef..23f125b 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which, u32 old_feature, u32 new_feature
 {
 	struct audit_buffer *ab;
 
-	if (audit_enabled == AUDIT_OFF)
+	if (!audit_enabled)
 		return;
-
-	ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
+	ab = audit_log_start(current->audit_context, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
 	if (!ab)
 		return;
 	audit_log_task_info(ab, current);
-- 
1.8.3.1

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Paul Moore]

On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
> Tie syscall information to FEATURE_CHANGE calls since it is a result of
> user action.
>
> See: https://github.com/linux-audit/audit-kernel/issues/80
>
> Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
> ---
>  kernel/audit.c | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/audit.c b/kernel/audit.c
> index 8da24ef..23f125b 100644
> --- a/kernel/audit.c
> +++ b/kernel/audit.c
> @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which, u32 old_feature, u32 new_feature
>  {
>         struct audit_buffer *ab;
>
> -       if (audit_enabled == AUDIT_OFF)
> +       if (!audit_enabled)

Sooo, this is an unrelated style change, why?  Looking at the rest of
kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
why are you adding noise to this patch?

>                 return;
> -
> -       ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
> +       ab = audit_log_start(current->audit_context, GFP_KERNEL, AUDIT_FEATURE_CHANGE);

This is the important part, and the Right Thing To Do.

>         if (!ab)
>                 return;
>         audit_log_task_info(ab, current);
> --
> 1.8.3.1

-- 
paul moore
www.paul-moore.com

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Steve Grubb]

On Tuesday, April 17, 2018 6:06:24 PM EDT Paul Moore wrote:
> On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
> > Tie syscall information to FEATURE_CHANGE calls since it is a result of
> > user action.
> > 
> > See: https://github.com/linux-audit/audit-kernel/issues/80
> > 
> > Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
> > ---
> > 
> >  kernel/audit.c | 5 ++---
> >  1 file changed, 2 insertions(+), 3 deletions(-)
> > 
> > diff --git a/kernel/audit.c b/kernel/audit.c
> > index 8da24ef..23f125b 100644
> > --- a/kernel/audit.c
> > +++ b/kernel/audit.c
> > @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which,
> > u32 old_feature, u32 new_feature> 
> >  {
> >  
> >         struct audit_buffer *ab;
> > 
> > -       if (audit_enabled == AUDIT_OFF)
> > +       if (!audit_enabled)
> 
> Sooo, this is an unrelated style change, why?  Looking at the rest of
> kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
> why are you adding noise to this patch?
> 
> >                 return;
> > 
> > -
> > -       ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
> > +       ab = audit_log_start(current->audit_context, GFP_KERNEL,
> > AUDIT_FEATURE_CHANGE);
> This is the important part, and the Right Thing To Do.

This is an unexpected change. I have asked questions on the github issue 
tracker but have not gotten a satisfactory answer. Please do not merge this 
until there's agreement on this.

Thanks,
-Steve

> >         if (!ab)
> >         
> >                 return;
> >         
> >         audit_log_task_info(ab, current);
> > 
> > --
> > 1.8.3.1

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Steve Grubb]

On Tuesday, April 17, 2018 6:06:24 PM EDT Paul Moore wrote:
> On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
> > Tie syscall information to FEATURE_CHANGE calls since it is a result of
> > user action.
> > 
> > See: https://github.com/linux-audit/audit-kernel/issues/80
> > 
> > Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
> > ---
> > 
> >  kernel/audit.c | 5 ++---
> >  1 file changed, 2 insertions(+), 3 deletions(-)
> > 
> > diff --git a/kernel/audit.c b/kernel/audit.c
> > index 8da24ef..23f125b 100644
> > --- a/kernel/audit.c
> > +++ b/kernel/audit.c
> > @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which,
> > u32 old_feature, u32 new_feature> 
> >  {
> >  
> >         struct audit_buffer *ab;
> > 
> > -       if (audit_enabled == AUDIT_OFF)
> > +       if (!audit_enabled)
> 
> Sooo, this is an unrelated style change, why?  Looking at the rest of
> kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
> why are you adding noise to this patch?
> 
> >                 return;
> > 
> > -
> > -       ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
> > +       ab = audit_log_start(current->audit_context, GFP_KERNEL,
> > AUDIT_FEATURE_CHANGE);
> This is the important part, and the Right Thing To Do.

This is an unexpected change. I have asked questions on the github issue 
tracker but have not gotten a satisfactory answer. Please do not merge this 
until there's agreement on this.

Thanks,
-Steve

> >         if (!ab)
> >         
> >                 return;
> >         
> >         audit_log_task_info(ab, current);
> > 
> > --
> > 1.8.3.1

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Paul Moore]

On Tue, Apr 17, 2018 at 6:10 PM, Steve Grubb <sgrubb@redhat.com> wrote:
> On Tuesday, April 17, 2018 6:06:24 PM EDT Paul Moore wrote:
>> On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
>> > Tie syscall information to FEATURE_CHANGE calls since it is a result of
>> > user action.
>> >
>> > See: https://github.com/linux-audit/audit-kernel/issues/80
>> >
>> > Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
>> > ---
>> >
>> >  kernel/audit.c | 5 ++---
>> >  1 file changed, 2 insertions(+), 3 deletions(-)
>> >
>> > diff --git a/kernel/audit.c b/kernel/audit.c
>> > index 8da24ef..23f125b 100644
>> > --- a/kernel/audit.c
>> > +++ b/kernel/audit.c
>> > @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which,
>> > u32 old_feature, u32 new_feature>
>> >  {
>> >
>> >         struct audit_buffer *ab;
>> >
>> > -       if (audit_enabled == AUDIT_OFF)
>> > +       if (!audit_enabled)
>>
>> Sooo, this is an unrelated style change, why?  Looking at the rest of
>> kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
>> why are you adding noise to this patch?
>>
>> >                 return;
>> >
>> > -
>> > -       ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
>> > +       ab = audit_log_start(current->audit_context, GFP_KERNEL,
>> > AUDIT_FEATURE_CHANGE);
>> This is the important part, and the Right Thing To Do.
>
> This is an unexpected change. I have asked questions on the github issue
> tracker but have not gotten a satisfactory answer. Please do not merge this
> until there's agreement on this.

It shouldn't be surprising, we've been talking about connecting
records for some time now, in different contexts and both on and off
list.  Not only does it helps pave the way for the audit container ID
work, it just makes sense.  I've seen your questions in this
particular GitHub issue and I think Richard has answered them
satisfactorily.

Once Richard removes the style change, or provides a good enough
reason for why it should stay in this patch, I plan on merging this
into audit/next.

-- 
paul moore
www.paul-moore.com

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Richard Guy Briggs]

On 2018-04-17 18:06, Paul Moore wrote:
> On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
> > Tie syscall information to FEATURE_CHANGE calls since it is a result of
> > user action.
> >
> > See: https://github.com/linux-audit/audit-kernel/issues/80
> >
> > Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
> > ---
> >  kernel/audit.c | 5 ++---
> >  1 file changed, 2 insertions(+), 3 deletions(-)
> >
> > diff --git a/kernel/audit.c b/kernel/audit.c
> > index 8da24ef..23f125b 100644
> > --- a/kernel/audit.c
> > +++ b/kernel/audit.c
> > @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which, u32 old_feature, u32 new_feature
> >  {
> >         struct audit_buffer *ab;
> >
> > -       if (audit_enabled == AUDIT_OFF)
> > +       if (!audit_enabled)
> 
> Sooo, this is an unrelated style change, why?  Looking at the rest of
> kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
> why are you adding noise to this patch?

Ok, survey sez 25 instances of audit_enabled used as a boolean vs 7
instances where it could be used as a boolean where the expression is
made harder to read (in my opinion).  I thought it was worth changing to
read the same way most of the other instances I've been reviewing are
written.  There are only two where the non-boolean distiction with
AUDIT_LOCKED is required.

> >                 return;
> > -
> > -       ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
> > +       ab = audit_log_start(current->audit_context, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
> 
> This is the important part, and the Right Thing To Do.
> 
> >         if (!ab)
> >                 return;
> >         audit_log_task_info(ab, current);
> 
> paul moore

- RGB

--
Richard Guy Briggs <rgb@redhat.com>
Sr. S/W Engineer, Kernel Security, Base Operating Systems
Remote, Ottawa, Red Hat Canada
IRC: rgb, SunRaycer
Voice: +1.647.777.2635, Internal: (81) 32635

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Paul Moore]

On Fri, Apr 20, 2018 at 9:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
> On 2018-04-17 18:06, Paul Moore wrote:
>> On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
>> > Tie syscall information to FEATURE_CHANGE calls since it is a result of
>> > user action.
>> >
>> > See: https://github.com/linux-audit/audit-kernel/issues/80
>> >
>> > Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
>> > ---
>> >  kernel/audit.c | 5 ++---
>> >  1 file changed, 2 insertions(+), 3 deletions(-)
>> >
>> > diff --git a/kernel/audit.c b/kernel/audit.c
>> > index 8da24ef..23f125b 100644
>> > --- a/kernel/audit.c
>> > +++ b/kernel/audit.c
>> > @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which, u32 old_feature, u32 new_feature
>> >  {
>> >         struct audit_buffer *ab;
>> >
>> > -       if (audit_enabled == AUDIT_OFF)
>> > +       if (!audit_enabled)
>>
>> Sooo, this is an unrelated style change, why?  Looking at the rest of
>> kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
>> why are you adding noise to this patch?
>
> Ok, survey sez 25 instances of audit_enabled used as a boolean vs 7
> instances where it could be used as a boolean where the expression is
> made harder to read (in my opinion).  I thought it was worth changing to
> read the same way most of the other instances I've been reviewing are
> written.  There are only two where the non-boolean distiction with
> AUDIT_LOCKED is required.

Thanks for the explanation.

While I still believe this patch, and connecting records in general,
is the Right Thing To Do, I'm expecting there to be some hate mail on
this issue and I would like to keep the patch as small and as
straight-to-the-point as possible just so there is little confusion
about what is changing.

Please respin this without the style change and I'll merge it as soon
as I see it.  Alternatively, give me the "ok" and I'll merge the patch
now and just drop the style change; after all we're talking about one
line in a five line patch ;)

>> >                 return;
>> > -
>> > -       ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
>> > +       ab = audit_log_start(current->audit_context, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
>>
>> This is the important part, and the Right Thing To Do.
>>
>> >         if (!ab)
>> >                 return;
>> >         audit_log_task_info(ab, current);

-- 
paul moore
www.paul-moore.com

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Richard Guy Briggs]

On 2018-04-20 11:58, Paul Moore wrote:
> On Fri, Apr 20, 2018 at 9:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
> > On 2018-04-17 18:06, Paul Moore wrote:
> >> On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
> >> > Tie syscall information to FEATURE_CHANGE calls since it is a result of
> >> > user action.
> >> >
> >> > See: https://github.com/linux-audit/audit-kernel/issues/80
> >> >
> >> > Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
> >> > ---
> >> >  kernel/audit.c | 5 ++---
> >> >  1 file changed, 2 insertions(+), 3 deletions(-)
> >> >
> >> > diff --git a/kernel/audit.c b/kernel/audit.c
> >> > index 8da24ef..23f125b 100644
> >> > --- a/kernel/audit.c
> >> > +++ b/kernel/audit.c
> >> > @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which, u32 old_feature, u32 new_feature
> >> >  {
> >> >         struct audit_buffer *ab;
> >> >
> >> > -       if (audit_enabled == AUDIT_OFF)
> >> > +       if (!audit_enabled)
> >>
> >> Sooo, this is an unrelated style change, why?  Looking at the rest of
> >> kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
> >> why are you adding noise to this patch?
> >
> > Ok, survey sez 25 instances of audit_enabled used as a boolean vs 7
> > instances where it could be used as a boolean where the expression is
> > made harder to read (in my opinion).  I thought it was worth changing to
> > read the same way most of the other instances I've been reviewing are
> > written.  There are only two where the non-boolean distiction with
> > AUDIT_LOCKED is required.
> 
> Thanks for the explanation.
> 
> While I still believe this patch, and connecting records in general,
> is the Right Thing To Do, I'm expecting there to be some hate mail on
> this issue and I would like to keep the patch as small and as
> straight-to-the-point as possible just so there is little confusion
> about what is changing.
> 
> Please respin this without the style change and I'll merge it as soon
> as I see it.  Alternatively, give me the "ok" and I'll merge the patch
> now and just drop the style change; after all we're talking about one
> line in a five line patch ;)

Go ahead and drop that style change line to simplify this patch and I'll
submit another patch to clean them all up at the same time (probably the
next time one of those changes).

> >> >                 return;
> >> > -
> >> > -       ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
> >> > +       ab = audit_log_start(current->audit_context, GFP_KERNEL, AUDIT_FEATURE_CHANGE);
> >>
> >> This is the important part, and the Right Thing To Do.
> >>
> >> >         if (!ab)
> >> >                 return;
> >> >         audit_log_task_info(ab, current);
> 
> -- 
> paul moore
> www.paul-moore.com

- RGB

--
Richard Guy Briggs <rgb@redhat.com>
Sr. S/W Engineer, Kernel Security, Base Operating Systems
Remote, Ottawa, Red Hat Canada
IRC: rgb, SunRaycer
Voice: +1.647.777.2635, Internal: (81) 32635

Re: [PATCH ghak80 V1] audit: add syscall information to FEATURE_CHANGE records

[Paul Moore]

On Fri, Apr 20, 2018 at 1:48 PM, Richard Guy Briggs <rgb@redhat.com> wrote:
> On 2018-04-20 11:58, Paul Moore wrote:
>> On Fri, Apr 20, 2018 at 9:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
>> > On 2018-04-17 18:06, Paul Moore wrote:
>> >> On Wed, Apr 11, 2018 at 8:46 AM, Richard Guy Briggs <rgb@redhat.com> wrote:
>> >> > Tie syscall information to FEATURE_CHANGE calls since it is a result of
>> >> > user action.
>> >> >
>> >> > See: https://github.com/linux-audit/audit-kernel/issues/80
>> >> >
>> >> > Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
>> >> > ---
>> >> >  kernel/audit.c | 5 ++---
>> >> >  1 file changed, 2 insertions(+), 3 deletions(-)
>> >> >
>> >> > diff --git a/kernel/audit.c b/kernel/audit.c
>> >> > index 8da24ef..23f125b 100644
>> >> > --- a/kernel/audit.c
>> >> > +++ b/kernel/audit.c
>> >> > @@ -1103,10 +1103,9 @@ static void audit_log_feature_change(int which, u32 old_feature, u32 new_feature
>> >> >  {
>> >> >         struct audit_buffer *ab;
>> >> >
>> >> > -       if (audit_enabled == AUDIT_OFF)
>> >> > +       if (!audit_enabled)
>> >>
>> >> Sooo, this is an unrelated style change, why?  Looking at the rest of
>> >> kernel/audit.c we seem to use a mix of "(!x)" and "(x == 0/CONST)" so
>> >> why are you adding noise to this patch?
>> >
>> > Ok, survey sez 25 instances of audit_enabled used as a boolean vs 7
>> > instances where it could be used as a boolean where the expression is
>> > made harder to read (in my opinion).  I thought it was worth changing to
>> > read the same way most of the other instances I've been reviewing are
>> > written.  There are only two where the non-boolean distiction with
>> > AUDIT_LOCKED is required.
>>
>> Thanks for the explanation.
>>
>> While I still believe this patch, and connecting records in general,
>> is the Right Thing To Do, I'm expecting there to be some hate mail on
>> this issue and I would like to keep the patch as small and as
>> straight-to-the-point as possible just so there is little confusion
>> about what is changing.
>>
>> Please respin this without the style change and I'll merge it as soon
>> as I see it.  Alternatively, give me the "ok" and I'll merge the patch
>> now and just drop the style change; after all we're talking about one
>> line in a five line patch ;)
>
> Go ahead and drop that style change line to simplify this patch and I'll
> submit another patch to clean them all up at the same time (probably the
> next time one of those changes).

Sounds good.

Merged.  Thanks Richard.

-- 
paul moore
www.paul-moore.com