Add __rcu annotation to RCU-protected global pointer auditd_conn.
auditd_conn is an RCU-protected global pointer,i.e., accessed
via RCU methods rcu_dereference() and rcu_assign_pointer(),
hence it must be annotated with __rcu for sparse to report
warnings/errors correctly.
Fix multiple instances of the sparse error:
error: incompatible types in comparison expression
(different address spaces)
Reviewed-by: Joel Fernandes (Google) <[email protected]>
Signed-off-by: Amol Grover <[email protected]>
---
v3:
- update changelog to be more descriptive
v2:
- fix erroneous RCU pointer initialization
kernel/audit.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/kernel/audit.c b/kernel/audit.c
index da8dc0db5bd3..ff7cfc61f53d 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -102,12 +102,13 @@ struct audit_net {
* This struct is RCU protected; you must either hold the RCU lock for reading
* or the associated spinlock for writing.
*/
-static struct auditd_connection {
+struct auditd_connection {
struct pid *pid;
u32 portid;
struct net *net;
struct rcu_head rcu;
-} *auditd_conn = NULL;
+};
+static struct auditd_connection __rcu *auditd_conn;
static DEFINE_SPINLOCK(auditd_conn_lock);
/* If audit_rate_limit is non-zero, limit the rate of sending audit records
--
2.24.0
Good idea to CC the following on RCU patches:
Paul McKenney
Steven Rostedt
(Any others on the RCU maintainers list).
And, the list: [email protected]
Could anyone Ack the patch? Looks safe and straight forward.
On Mon, Dec 02, 2019 at 12:03:48AM +0530, Amol Grover wrote:
> Add __rcu annotation to RCU-protected global pointer auditd_conn.
>
> auditd_conn is an RCU-protected global pointer,i.e., accessed
> via RCU methods rcu_dereference() and rcu_assign_pointer(),
> hence it must be annotated with __rcu for sparse to report
> warnings/errors correctly.
>
> Fix multiple instances of the sparse error:
> error: incompatible types in comparison expression
> (different address spaces)
>
> Reviewed-by: Joel Fernandes (Google) <[email protected]>
> Signed-off-by: Amol Grover <[email protected]>
> ---
> v3:
> - update changelog to be more descriptive
>
> v2:
> - fix erroneous RCU pointer initialization
>
> kernel/audit.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/kernel/audit.c b/kernel/audit.c
> index da8dc0db5bd3..ff7cfc61f53d 100644
> --- a/kernel/audit.c
> +++ b/kernel/audit.c
> @@ -102,12 +102,13 @@ struct audit_net {
> * This struct is RCU protected; you must either hold the RCU lock for reading
> * or the associated spinlock for writing.
> */
> -static struct auditd_connection {
> +struct auditd_connection {
> struct pid *pid;
> u32 portid;
> struct net *net;
> struct rcu_head rcu;
> -} *auditd_conn = NULL;
> +};
> +static struct auditd_connection __rcu *auditd_conn;
> static DEFINE_SPINLOCK(auditd_conn_lock);
>
> /* If audit_rate_limit is non-zero, limit the rate of sending audit records
> --
> 2.24.0
>
On Mon, 2 Dec 2019 16:19:15 -0500
Joel Fernandes <[email protected]> wrote:
> Good idea to CC the following on RCU patches:
> Paul McKenney
> Steven Rostedt
I'm fine with this if it doesn't cause any rcu splats with sparse. Not
sure if use cases of RCU requires RCU maintainers Cc'd. Although we can
usually keep people from misusing it ;-)
-- Steve
> (Any others on the RCU maintainers list).
> And, the list: [email protected]
>
> Could anyone Ack the patch? Looks safe and straight forward.
>
> On Mon, Dec 02, 2019 at 12:03:48AM +0530, Amol Grover wrote:
> > Add __rcu annotation to RCU-protected global pointer auditd_conn.
> >
> > auditd_conn is an RCU-protected global pointer,i.e., accessed
> > via RCU methods rcu_dereference() and rcu_assign_pointer(),
> > hence it must be annotated with __rcu for sparse to report
> > warnings/errors correctly.
> >
> > Fix multiple instances of the sparse error:
> > error: incompatible types in comparison expression
> > (different address spaces)
> >
> > Reviewed-by: Joel Fernandes (Google) <[email protected]>
> > Signed-off-by: Amol Grover <[email protected]>
> > ---
> > v3:
> > - update changelog to be more descriptive
> >
> > v2:
> > - fix erroneous RCU pointer initialization
> >
> > kernel/audit.c | 5 +++--
> > 1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/kernel/audit.c b/kernel/audit.c
> > index da8dc0db5bd3..ff7cfc61f53d 100644
> > --- a/kernel/audit.c
> > +++ b/kernel/audit.c
> > @@ -102,12 +102,13 @@ struct audit_net {
> > * This struct is RCU protected; you must either hold the RCU lock for reading
> > * or the associated spinlock for writing.
> > */
> > -static struct auditd_connection {
> > +struct auditd_connection {
> > struct pid *pid;
> > u32 portid;
> > struct net *net;
> > struct rcu_head rcu;
> > -} *auditd_conn = NULL;
> > +};
> > +static struct auditd_connection __rcu *auditd_conn;
> > static DEFINE_SPINLOCK(auditd_conn_lock);
> >
> > /* If audit_rate_limit is non-zero, limit the rate of sending audit records
> > --
> > 2.24.0
> >
On Mon, Dec 2, 2019 at 4:19 PM Joel Fernandes <[email protected]> wrote:
> Good idea to CC the following on RCU patches:
> Paul McKenney
> Steven Rostedt
> (Any others on the RCU maintainers list).
> And, the list: [email protected]
>
> Could anyone Ack the patch? Looks safe and straight forward.
FWIW, this looks reasonable to me, but I don't see this as a critical
fix that needs to go in during the merge window. Unless I see any
objections, I'll plan on merging this into audit/next once the merge
window closes.
> On Mon, Dec 02, 2019 at 12:03:48AM +0530, Amol Grover wrote:
> > Add __rcu annotation to RCU-protected global pointer auditd_conn.
> >
> > auditd_conn is an RCU-protected global pointer,i.e., accessed
> > via RCU methods rcu_dereference() and rcu_assign_pointer(),
> > hence it must be annotated with __rcu for sparse to report
> > warnings/errors correctly.
> >
> > Fix multiple instances of the sparse error:
> > error: incompatible types in comparison expression
> > (different address spaces)
> >
> > Reviewed-by: Joel Fernandes (Google) <[email protected]>
> > Signed-off-by: Amol Grover <[email protected]>
> > ---
> > v3:
> > - update changelog to be more descriptive
> >
> > v2:
> > - fix erroneous RCU pointer initialization
> >
> > kernel/audit.c | 5 +++--
> > 1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/kernel/audit.c b/kernel/audit.c
> > index da8dc0db5bd3..ff7cfc61f53d 100644
> > --- a/kernel/audit.c
> > +++ b/kernel/audit.c
> > @@ -102,12 +102,13 @@ struct audit_net {
> > * This struct is RCU protected; you must either hold the RCU lock for reading
> > * or the associated spinlock for writing.
> > */
> > -static struct auditd_connection {
> > +struct auditd_connection {
> > struct pid *pid;
> > u32 portid;
> > struct net *net;
> > struct rcu_head rcu;
> > -} *auditd_conn = NULL;
> > +};
> > +static struct auditd_connection __rcu *auditd_conn;
> > static DEFINE_SPINLOCK(auditd_conn_lock);
> >
> > /* If audit_rate_limit is non-zero, limit the rate of sending audit records
> > --
> > 2.24.0
> >
--
paul moore
http://www.paul-moore.com
On Mon, Dec 02, 2019 at 06:24:29PM -0500, Paul Moore wrote:
> On Mon, Dec 2, 2019 at 4:19 PM Joel Fernandes <[email protected]> wrote:
> > Good idea to CC the following on RCU patches:
> > Paul McKenney
> > Steven Rostedt
> > (Any others on the RCU maintainers list).
> > And, the list: [email protected]
> >
> > Could anyone Ack the patch? Looks safe and straight forward.
>
> FWIW, this looks reasonable to me, but I don't see this as a critical
> fix that needs to go in during the merge window. Unless I see any
> objections, I'll plan on merging this into audit/next once the merge
> window closes.
Sounds good, thanks!
- Joel
> > On Mon, Dec 02, 2019 at 12:03:48AM +0530, Amol Grover wrote:
> > > Add __rcu annotation to RCU-protected global pointer auditd_conn.
> > >
> > > auditd_conn is an RCU-protected global pointer,i.e., accessed
> > > via RCU methods rcu_dereference() and rcu_assign_pointer(),
> > > hence it must be annotated with __rcu for sparse to report
> > > warnings/errors correctly.
> > >
> > > Fix multiple instances of the sparse error:
> > > error: incompatible types in comparison expression
> > > (different address spaces)
> > >
> > > Reviewed-by: Joel Fernandes (Google) <[email protected]>
> > > Signed-off-by: Amol Grover <[email protected]>
> > > ---
> > > v3:
> > > - update changelog to be more descriptive
> > >
> > > v2:
> > > - fix erroneous RCU pointer initialization
> > >
> > > kernel/audit.c | 5 +++--
> > > 1 file changed, 3 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/kernel/audit.c b/kernel/audit.c
> > > index da8dc0db5bd3..ff7cfc61f53d 100644
> > > --- a/kernel/audit.c
> > > +++ b/kernel/audit.c
> > > @@ -102,12 +102,13 @@ struct audit_net {
> > > * This struct is RCU protected; you must either hold the RCU lock for reading
> > > * or the associated spinlock for writing.
> > > */
> > > -static struct auditd_connection {
> > > +struct auditd_connection {
> > > struct pid *pid;
> > > u32 portid;
> > > struct net *net;
> > > struct rcu_head rcu;
> > > -} *auditd_conn = NULL;
> > > +};
> > > +static struct auditd_connection __rcu *auditd_conn;
> > > static DEFINE_SPINLOCK(auditd_conn_lock);
> > >
> > > /* If audit_rate_limit is non-zero, limit the rate of sending audit records
> > > --
> > > 2.24.0
> > >
>
>
>
> --
> paul moore
> http://www.paul-moore.com
On Mon, Dec 02, 2019 at 05:26:39PM -0500, Steven Rostedt wrote:
> On Mon, 2 Dec 2019 16:19:15 -0500
> Joel Fernandes <[email protected]> wrote:
>
> > Good idea to CC the following on RCU patches:
> > Paul McKenney
> > Steven Rostedt
>
> I'm fine with this if it doesn't cause any rcu splats with sparse. Not
> sure if use cases of RCU requires RCU maintainers Cc'd. Although we can
> usually keep people from misusing it ;-)
Yes true :) Thanks for taking a look.
- Joel
>
> -- Steve
>
>
> > (Any others on the RCU maintainers list).
> > And, the list: [email protected]
> >
> > Could anyone Ack the patch? Looks safe and straight forward.
> >
> > On Mon, Dec 02, 2019 at 12:03:48AM +0530, Amol Grover wrote:
> > > Add __rcu annotation to RCU-protected global pointer auditd_conn.
> > >
> > > auditd_conn is an RCU-protected global pointer,i.e., accessed
> > > via RCU methods rcu_dereference() and rcu_assign_pointer(),
> > > hence it must be annotated with __rcu for sparse to report
> > > warnings/errors correctly.
> > >
> > > Fix multiple instances of the sparse error:
> > > error: incompatible types in comparison expression
> > > (different address spaces)
> > >
> > > Reviewed-by: Joel Fernandes (Google) <[email protected]>
> > > Signed-off-by: Amol Grover <[email protected]>
> > > ---
> > > v3:
> > > - update changelog to be more descriptive
> > >
> > > v2:
> > > - fix erroneous RCU pointer initialization
> > >
> > > kernel/audit.c | 5 +++--
> > > 1 file changed, 3 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/kernel/audit.c b/kernel/audit.c
> > > index da8dc0db5bd3..ff7cfc61f53d 100644
> > > --- a/kernel/audit.c
> > > +++ b/kernel/audit.c
> > > @@ -102,12 +102,13 @@ struct audit_net {
> > > * This struct is RCU protected; you must either hold the RCU lock for reading
> > > * or the associated spinlock for writing.
> > > */
> > > -static struct auditd_connection {
> > > +struct auditd_connection {
> > > struct pid *pid;
> > > u32 portid;
> > > struct net *net;
> > > struct rcu_head rcu;
> > > -} *auditd_conn = NULL;
> > > +};
> > > +static struct auditd_connection __rcu *auditd_conn;
> > > static DEFINE_SPINLOCK(auditd_conn_lock);
> > >
> > > /* If audit_rate_limit is non-zero, limit the rate of sending audit records
> > > --
> > > 2.24.0
> > >
>
On Mon, Dec 2, 2019 at 6:35 PM Joel Fernandes <[email protected]> wrote:
> On Mon, Dec 02, 2019 at 06:24:29PM -0500, Paul Moore wrote:
> > On Mon, Dec 2, 2019 at 4:19 PM Joel Fernandes <[email protected]> wrote:
> > > Good idea to CC the following on RCU patches:
> > > Paul McKenney
> > > Steven Rostedt
> > > (Any others on the RCU maintainers list).
> > > And, the list: [email protected]
> > >
> > > Could anyone Ack the patch? Looks safe and straight forward.
> >
> > FWIW, this looks reasonable to me, but I don't see this as a critical
> > fix that needs to go in during the merge window. Unless I see any
> > objections, I'll plan on merging this into audit/next once the merge
> > window closes.
>
> Sounds good, thanks!
FYI, it's in audit/next now. Thanks again.
--
paul moore
http://www.paul-moore.com