From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Howells Subject: Re: [PATCH net-next] udp: Add tracepoints to monitor skbs going in and out of a UDP socket Date: Fri, 05 Oct 2018 07:46:50 +0100 Message-ID: <9724.1538722010@warthog.procyon.org.uk> References: <20181005020316.7cqlcfkedhamwhtt@ast-mbp.dhcp.thefacebook.com> <20181004171925.6neebizw4xmpjhv6@ast-mbp.dhcp.thefacebook.com> <153864184110.20296.16498481420320274305.stgit@warthog.procyon.org.uk> <25668.1538672073@warthog.procyon.org.uk> <5213.1538695482@warthog.procyon.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Cc: dhowells@redhat.com, David Ahern , netdev@vger.kernel.org, linux-afs@lists.infradead.org, linux-kernel@vger.kernel.org To: Alexei Starovoitov Return-path: In-Reply-To: <20181005020316.7cqlcfkedhamwhtt@ast-mbp.dhcp.thefacebook.com> Content-ID: <9723.1538722010.1@warthog.procyon.org.uk> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Alexei Starovoitov wrote: > - kprobes at the top of the function don't use traps and they've been > optimized over the years to have very low overhead To quote kprobes.txt: How Does a Kprobe Work? ----------------------- When a kprobe is registered, Kprobes makes a copy of the probed instruction and replaces the first byte(s) of the probed instruction with a breakpoint instruction (e.g., int3 on i386 and x86_64). Perhaps the docs need updating. However, for my purposes, tracepoints are easier. Anyway, as I said, I don't feel that strongly about the patch. It was useful for me, and I thought it might be useful for other people. David