Hi Richard, > -----Original Message----- > From: Richard Cochran [mailto:richardcoch...@gmail.com] > Sent: Tuesday, June 5, 2018 9:58 PM > To: Y.b. Lu <yangbo...@nxp.com> > Cc: net...@vger.kernel.org; Madalin-cristian Bucur > <madalin.bu...@nxp.com>; Rob Herring <robh...@kernel.org>; Shawn Guo > <shawn...@kernel.org>; David S . Miller <da...@davemloft.net>; > devicet...@vger.kernel.org; linuxppc-dev@lists.ozlabs.org; > linux-arm-ker...@lists.infradead.org; linux-ker...@vger.kernel.org > Subject: Re: [PATCH 09/10] dpaa_eth: add support for hardware timestamping > > On Tue, Jun 05, 2018 at 03:35:28AM +0000, Y.b. Lu wrote: > > [Y.b. Lu] Actually these timestamping codes affected DPAA networking > performance in our previous performance test. > > That's why we used ifdef for it. > > How much does time stamping hurt performance? > > If the time stamping is compiled in but not enabled at run time, does it still > affect performace?
[Y.b. Lu] I can't remember and find the old data since it had been a long time. I just did the iperf test today between two 10G ports. I didn’t see any performance changes with timestamping code 😊 So, let's me remove the ifdef in next version. Thanks a lot. > > Thanks, > Richard
