> -----Original Message----- > From: Yuanhan Liu [mailto:yuanhan....@linux.intel.com] > Sent: Wednesday, December 7, 2016 10:19 AM > To: Ananyev, Konstantin <konstantin.anan...@intel.com> > Cc: Yigit, Ferruh <ferruh.yi...@intel.com>; Olivier Matz > <olivier.m...@6wind.com>; Thomas Monjalon <thomas.monja...@6wind.com>; > dev@dpdk.org; Jan Medala <j...@semihalf.com>; Jakub Palider > <j...@semihalf.com>; Netanel Belgazal <neta...@amazon.com>; Evgeny > Schemeilin <evge...@amazon.com>; Alejandro Lucero > <alejandro.luc...@netronome.com>; Yong Wang <yongw...@vmware.com>; > Andrew Rybchenko <arybche...@solarflare.com>; Hemant Agrawal > <hemant.agra...@nxp.com>; Kulasek, TomaszX > <tomaszx.kula...@intel.com> > Subject: Re: [dpdk-dev] [PATCH v12 0/6] add Tx preparation > > On Wed, Dec 07, 2016 at 10:13:14AM +0000, Ananyev, Konstantin wrote: > > > > Hi Yliu, > > > > > > > > On Tue, Dec 06, 2016 at 03:53:42PM +0000, Ferruh Yigit wrote: > > > > > Please, we need a comment for each driver saying > > > > > "it is OK, we do not need any checksum preparation for TSO" > > > > > or > > > > > "yes we have to implement tx_prepare or TSO will not work in this > > > > > mode" > > > > > > > > > > > Sorry for late. For virtio, I think it's not a must. The checksum stuff > > > has been handled inside the Tx function. However, we may could move it > > > to tx_prepare, which would actually recover the performance lost > > > introduced while enabling TSO for the non-TSO case. > > > > > > > So would you like to provide a patch for it, > > Or would you like to keep tx_prepare() for virtio as NOP for now? > > Hi Konstantin, > > I'd keep it as it is for now. It should be a trivial patch after all, that > I could provide it when everything are settled down.
Ok, thanks for clarification. Konstantin > > --yliu
