From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steven Rostedt Subject: Re: [PATCH net-next] ipv6: let trace_fib6_table_lookup() dereference the fib table Date: Thu, 19 Oct 2017 09:30:25 -0400 Message-ID: <20171019093025.5a7b9ea2@vmware.local.home> References: <48235df8fdf787356070d8baa1ac9816ef5736d4.1508398120.git.pabeni@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, "David S. Miller" , Ingo Molnar , David Ahern To: Paolo Abeni Return-path: Received: from smtprelay0082.hostedemail.com ([216.40.44.82]:56583 "EHLO smtprelay.hostedemail.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752969AbdJSNa3 (ORCPT ); Thu, 19 Oct 2017 09:30:29 -0400 In-Reply-To: <48235df8fdf787356070d8baa1ac9816ef5736d4.1508398120.git.pabeni@redhat.com> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, 19 Oct 2017 09:31:43 +0200 Paolo Abeni wrote: > The perf traces for ipv6 routing code show a relevant cost around > trace_fib6_table_lookup(), even if no trace is enabled. This is > due to the fib6_table de-referencing currently performed by the > caller. > > Let's the tracing code pay this overhead, passing to the trace > helper the table pointer. This gives small but measurable > performance improvement under UDP flood. > > Signed-off-by: Paolo Abeni > --- Acked-by: Steven Rostedt (VMware) -- Steve