> > > I wonder if you could get much back, in terms of performance, by moving
> > > the
> > >          context->dummy = !audit_n_rules;
> > > line to the top and just returning if context->dummy == 1;
> > 
> > We get 668.09 cycles with this optimisation, so it comes down a bit, but
> > no where near if the auditing is disabled altogether.
> 
> Clean that patch up and send it.  Sounds like a win no matter what else
> we do.

ok...

audit: speedup audit_syscall_entry when there are zero rules

This creates a check at the start of audit_syscall_entry to see if there
are zero rules in the audit filter list.  If there are zero rules return
immediately.

This buys about ~10% on a null syscall on powerpc.

Signed-off-by: Michael Neuling <[email protected]>

diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 1b31c13..bc0872b 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -1579,6 +1579,9 @@ void audit_syscall_entry(int arch, int major,
        if (unlikely(!context))
                return;
 
+       context->dummy = !audit_n_rules;
+       if (context->dummy == 1)
+               return;
        /*
         * This happens only on certain architectures that make system
         * calls in kernel_thread via the entry.S interface, instead of
@@ -1628,7 +1631,6 @@ void audit_syscall_entry(int arch, int major,
        context->argv[3]    = a4;
 
        state = context->state;
-       context->dummy = !audit_n_rules;
        if (!context->dummy && state == AUDIT_BUILD_CONTEXT) {
                context->prio = 0;
                state = audit_filter_syscall(tsk, context, 
&audit_filter_list[AUDIT_FILTER_ENTRY]);

--
Linux-audit mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/linux-audit

Reply via email to