From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steve Grubb Subject: Re: Watch Performance Date: Sun, 9 Apr 2006 15:48:41 -0400 Message-ID: <200604091548.41324.sgrubb@redhat.com> References: <200604081221.58080.sgrubb@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <200604081221.58080.sgrubb@redhat.com> Content-Disposition: inline List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-audit-bounces@redhat.com Errors-To: linux-audit-bounces@redhat.com To: linux-audit@redhat.com Cc: redhat-lspp@redhat.com List-Id: linux-audit@redhat.com Hi, Based on finding an unnecessary function call to selinux_task_ctxid when=20 evaluating syscall rules, I built a new kernel and re-ran the same tests. rules =A0seconds loss 0 =A0 =A0 =A0 =A047 0% 10 =A0 =A0 =A053 11% 25 =A0 =A0 =A068 43% 50 =A0 =A0 =A099 109% 75 =A0 =A0 =A0132 178% 90 =A0 =A0 =A0157 232% The 75 rule performance hit is now 178% instead of 184%. So there is some= =20 notable improvement in performance.=20 For comparison, I also loaded the 90 rules config into RHEL4. There is on= ly a=20 6% performance hit compared to no rules. I think the bulk of that comes f= rom=20 evaluating the 10 syscall rules rather than the file system audit code. -Steve