From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Cc: Thomas Monjalon <thomas.monjalon@6wind.com>,
"dev@dpdk.org" <dev@dpdk.org>,
Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>,
Stephen Hurd <stephen.hurd@broadcom.com>,
Jan Medala <jan@semihalf.com>, Jakub Palider <jpa@semihalf.com>,
John Daley <johndale@cisco.com>,
Alejandro Lucero <alejandro.lucero@netronome.com>,
Harish Patil <harish.patil@qlogic.com>,
Rasesh Mody <rasesh.mody@qlogic.com>,
Jerin Jacob <jerin.jacob@caviumnetworks.com>,
Yuanhan Liu <yuanhan.liu@linux.intel.com>,
Yong Wang <yongwang@vmware.com>,
"Kulasek, TomaszX" <tomaszx.kulasek@intel.com>,
"olivier.matz@6wind.com" <olivier.matz@6wind.com>
Subject: Re: [dpdk-dev] [PATCH v12 0/6] add Tx preparation
Date: Fri, 2 Dec 2016 01:00:55 +0000 [thread overview]
Message-ID: <2601191342CEEE43887BDE71AB9772583F0E2A9A@irsmsx105.ger.corp.intel.com> (raw)
In-Reply-To: <20161201071518.GG10340@6wind.com>
Hi Adrien,
>
> Hi Konstantin,
>
> On Wed, Nov 30, 2016 at 10:54:50AM +0000, Ananyev, Konstantin wrote:
> [...]
> > > Something is definitely needed here, and only PMDs can provide it. I think
> > > applications should not have to clear checksum fields or initialize them to
> > > some magic value, same goes for any other offload or hardware limitation
> > > that needs to be worked around.
> > >
> > > tx_prep() is one possible answer to this issue, however as mentioned in the
> > > original patch it can be very expensive if exposed by the PMD.
> > >
> > > Another issue I'm more concerned about is the way limitations are managed
> > > (struct rte_eth_desc_lim). While not officially tied to tx_prep(), this
> > > structure contains new fields that are only relevant to a few devices, and I
> > > fear it will keep growing with each new hardware quirk to manage, breaking
> > > ABIs in the process.
> >
> > Well, if some new HW capability/limitation would arise and we'd like to support
> > it in DPDK, then yes we probably would need to think how to incorporate it here.
> > Do you have anything particular in mind here?
>
> Nothing in particular, so for the sake of the argument, let's suppose that I
> would like to add a field to expose some limitation that only applies to my
> PMD during TX but looks generic enough to make sense, e.g. maximum packet
> size when VLAN tagging is requested.
Hmm, I didn't hear about such limitations so far, but if it is real case -
sure, feel free to submit the patch.
> PMDs are free to set that field to some
> special value (say, 0) if they do not care.
>
> Since that field exists however, conscious applications should check its
> value for each packet that needs to be transmitted. This extra code causes a
> slowdown just by sitting in the data path. Since it is not the only field in
> that structure, the performance impact can be significant.
>
> Even though this code is inside applications, it remains unfair to PMDs for
> which these tests are irrelevant. This problem is identified and addressed
> by tx_prepare().
I suppose the question is why do we need:
uint16_t nb_seg_max;
uint16_t nb_mtu_seg_max;
as we now have tx_prepare(), right?
For two reasons:
1. Some people might feel that tx_prepare() is not good (smart/fast) enough
for them and would prefer to do necessary preparations for TX offloads themselves.
2. Even if people do use tx_prepare() they still should take this information into accout.
As an example ixgbe can't TX packets with then 40 segments.
Obviously ixbge_tx_prep() performs that check and returns an error.
But it wouldn't try to merge/reallocate mbufs for you.
User still has to do it himself, or just prevent creating such long chains somehow.
>
> Thanks to tx_prepare(), these checks are moved back into PMDs where they
> belong. PMDs that do not need them do not have to provide support for
> tx_prepare() and do not suffer any performance impact as result;
> applications only have to make sure tx_prepare() is always called at some
> point before tx_burst().
>
> Once you reach this stage, you've effectively made tx_prepare() mandatory
> before tx_burst(). If some bug occurs, then perhaps you forgot to call
> tx_prepare(), you just need to add it. The total cost for doing TX is
> therefore tx_prepare() + tx_burst().
>
> I'm perhaps a bit pessimistic mind you, but I do not think tx_prepare() will
> remain optional for long. Sure, PMDs that do not implement it do not care,
> I'm focusing on applications, for which the performance impact of calling
> tx_prepare() followed by tx_burst() is higher than a single tx_burst()
> performing all the necessary preparation at once.
>
> [...]
> > > Following the same logic, why can't such a thing be made part of the TX
> > > burst function as well (through a direct call to rte_phdr_cksum_fix()
> > > whenever necessary). From an application standpoint, what are the advantages
> > > of having to:
> > >
> > > if (tx_prep()) // iterate and update mbufs as needed
> > > tx_burst(); // iterate and send
> > >
> > > Compared to:
> > >
> > > tx_burst(); // iterate, update as needed and send
> >
> > I think that was discussed extensively quite a lot previously here:
> > As Thomas already replied - main motivation is to allow user
> > to execute them on different stages of packet TX pipeline,
> > and probably on different cores.
> > I think that provides better flexibility to the user to when/where
> > do these preparations and hopefully would lead to better performance.
>
> And I agree, I think this use case is valid but does not warrant such a high
> penalty when your application does not need that much flexibility. Simple
> (yet conscious) applications need the highest performance. Complex ones as
> you described already suffer quite a bit from IPCs and won't mind a couple
> of extra CPU cycles right?
It would mean an extra cache-miss for every packet, so I think performance hit
would be quite significant.
About the 'simple' case when tx_prep() and tx_burst() are called on the same core,
Why do you believe that:
tx_prep(); tx_burst(); would be much slower than tx_burst() {tx_prep(), ...}?
tx_prep() itself is quite expensive, let say for Intel HW it includes:
- read mbuf fileds (2 cache-lines),
- read packet header (1/2 cache-lines)
- calculate pseudo-header csum
- update packet header
Comparing to that price of extra function call seems neglectable
(if we TX packets in bursts of course).
>
> Yes they will, therefore we need a method that satisfies both cases.
>
> As a possible solution, a special mbuf flag could be added to each mbuf
> having gone through tx_prepare(). That way, tx_burst() could skip some
> checks and things it would otherwise have done.
That's an interesting idea, but it has one drawback:
As I understand, it means that from now on if user doing preparations on his own,
he had to setup this flag, otherwise tx_burst() would do extra unnecessary work.
So any existing applications that using TX offloads and do preparation by themselves
would have to be modified to avoid performance loss.
>
> Another possibility, telling the PMD first that you always intend to use
> tx_prepare() and getting a simpler/faster tx_burst() callback as a result.
That what we have right now (at least for Intel HW):
it is a user responsibility to do the necessary preparations/checks before calling tx_burst().
With tx_prepare() we just remove from user the headache to implement tx_prepare() on his own.
Now he can use a 'proper' PMD provided function.
My vote still would be for that model.
Konstantin
next prev parent reply other threads:[~2016-12-02 1:01 UTC|newest]
Thread overview: 260+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-26 16:22 [dpdk-dev] [PATCH " Tomasz Kulasek
2016-08-26 16:22 ` [dpdk-dev] [PATCH 1/6] ethdev: " Tomasz Kulasek
2016-09-08 7:28 ` Jerin Jacob
2016-09-08 16:09 ` Kulasek, TomaszX
2016-09-09 5:58 ` Jerin Jacob
2016-08-26 16:22 ` [dpdk-dev] [PATCH 2/6] e1000: " Tomasz Kulasek
2016-08-26 16:22 ` [dpdk-dev] [PATCH 3/6] fm10k: " Tomasz Kulasek
2016-08-26 16:22 ` [dpdk-dev] [PATCH 4/6] i40e: " Tomasz Kulasek
2016-08-26 16:22 ` [dpdk-dev] [PATCH 5/6] ixgbe: " Tomasz Kulasek
2016-08-26 16:22 ` [dpdk-dev] [PATCH 6/6] testpmd: add txprep engine Tomasz Kulasek
2016-08-26 17:31 ` [dpdk-dev] [PATCH 0/6] add Tx preparation Stephen Hemminger
2016-08-31 12:34 ` Ananyev, Konstantin
2016-09-12 14:44 ` [dpdk-dev] [PATCH v2 " Tomasz Kulasek
2016-09-12 14:44 ` [dpdk-dev] [PATCH v2 1/6] ethdev: " Tomasz Kulasek
2016-09-19 13:03 ` Ananyev, Konstantin
2016-09-19 15:29 ` Kulasek, TomaszX
2016-09-19 16:06 ` Jerin Jacob
2016-09-20 9:06 ` Ananyev, Konstantin
2016-09-21 8:29 ` Jerin Jacob
2016-09-22 9:36 ` Ananyev, Konstantin
2016-09-22 9:59 ` Jerin Jacob
2016-09-23 9:41 ` Ananyev, Konstantin
2016-09-23 10:29 ` Jerin Jacob
2016-09-12 14:44 ` [dpdk-dev] [PATCH v2 2/6] e1000: " Tomasz Kulasek
2016-09-12 14:44 ` [dpdk-dev] [PATCH v2 3/6] fm10k: " Tomasz Kulasek
2016-09-12 14:44 ` [dpdk-dev] [PATCH v2 4/6] i40e: " Tomasz Kulasek
2016-09-12 14:44 ` [dpdk-dev] [PATCH v2 5/6] ixgbe: " Tomasz Kulasek
2016-09-19 12:54 ` Ananyev, Konstantin
2016-09-19 13:58 ` Kulasek, TomaszX
2016-09-19 15:23 ` Ananyev, Konstantin
2016-09-20 7:15 ` Ananyev, Konstantin
2016-09-12 14:44 ` [dpdk-dev] [PATCH v2 6/6] testpmd: add txprep engine Tomasz Kulasek
2016-09-19 12:59 ` Ananyev, Konstantin
2016-09-28 11:10 ` [dpdk-dev] [PATCH v3 0/6] add Tx preparation Tomasz Kulasek
2016-09-28 11:10 ` [dpdk-dev] [PATCH v3 1/6] ethdev: " Tomasz Kulasek
2016-09-29 10:40 ` Ananyev, Konstantin
2016-09-29 13:04 ` Kulasek, TomaszX
2016-09-29 13:57 ` Ananyev, Konstantin
2016-09-28 11:10 ` [dpdk-dev] [PATCH v3 2/6] e1000: " Tomasz Kulasek
2016-09-28 11:10 ` [dpdk-dev] [PATCH v3 3/6] fm10k: " Tomasz Kulasek
2016-09-28 11:10 ` [dpdk-dev] [PATCH v3 4/6] i40e: " Tomasz Kulasek
2016-09-28 11:10 ` [dpdk-dev] [PATCH v3 5/6] ixgbe: " Tomasz Kulasek
2016-09-29 11:09 ` Ananyev, Konstantin
2016-09-29 15:12 ` Kulasek, TomaszX
2016-09-29 17:01 ` Ananyev, Konstantin
2016-09-28 11:10 ` [dpdk-dev] [PATCH v3 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-09-30 9:00 ` [dpdk-dev] [PATCH v4 0/6] add Tx preparation Tomasz Kulasek
2016-09-30 9:00 ` [dpdk-dev] [PATCH v4 1/6] ethdev: " Tomasz Kulasek
2016-10-10 14:08 ` Thomas Monjalon
2016-10-13 7:08 ` Thomas Monjalon
2016-10-13 10:47 ` Kulasek, TomaszX
2016-09-30 9:00 ` [dpdk-dev] [PATCH v4 2/6] e1000: " Tomasz Kulasek
2016-09-30 9:00 ` [dpdk-dev] [PATCH v4 3/6] fm10k: " Tomasz Kulasek
2016-09-30 9:00 ` [dpdk-dev] [PATCH v4 4/6] i40e: " Tomasz Kulasek
2016-10-10 14:02 ` Wu, Jingjing
2016-10-10 17:20 ` Kulasek, TomaszX
2016-09-30 9:00 ` [dpdk-dev] [PATCH v4 5/6] ixgbe: " Tomasz Kulasek
2016-09-30 9:00 ` [dpdk-dev] [PATCH v4 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-09-30 9:55 ` [dpdk-dev] [PATCH v4 0/6] add Tx preparation Ananyev, Konstantin
2016-10-13 17:36 ` [dpdk-dev] [PATCH v5 " Tomasz Kulasek
2016-10-13 17:36 ` [dpdk-dev] [PATCH v5 1/6] ethdev: " Tomasz Kulasek
2016-10-13 19:21 ` Thomas Monjalon
2016-10-14 14:02 ` Kulasek, TomaszX
2016-10-14 14:20 ` Thomas Monjalon
2016-10-17 16:25 ` Kulasek, TomaszX
2016-10-13 17:36 ` [dpdk-dev] [PATCH v5 2/6] e1000: " Tomasz Kulasek
2016-10-13 17:36 ` [dpdk-dev] [PATCH v5 3/6] fm10k: " Tomasz Kulasek
2016-10-13 17:37 ` [dpdk-dev] [PATCH v5 4/6] i40e: " Tomasz Kulasek
2016-10-13 17:37 ` [dpdk-dev] [PATCH v5 5/6] ixgbe: " Tomasz Kulasek
2016-10-13 17:37 ` [dpdk-dev] [PATCH v5 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-10-14 15:05 ` [dpdk-dev] [PATCH v6 0/6] add Tx preparation Tomasz Kulasek
2016-10-14 15:05 ` [dpdk-dev] [PATCH v6 1/6] ethdev: " Tomasz Kulasek
2016-10-18 14:57 ` Olivier Matz
2016-10-19 15:42 ` Kulasek, TomaszX
2016-10-19 22:07 ` Ananyev, Konstantin
2016-10-14 15:05 ` [dpdk-dev] [PATCH v6 2/6] e1000: " Tomasz Kulasek
2016-10-14 15:05 ` [dpdk-dev] [PATCH v6 3/6] fm10k: " Tomasz Kulasek
2016-10-14 15:05 ` [dpdk-dev] [PATCH v6 4/6] i40e: " Tomasz Kulasek
2016-10-14 15:05 ` [dpdk-dev] [PATCH v6 5/6] ixgbe: " Tomasz Kulasek
2016-10-14 15:05 ` [dpdk-dev] [PATCH v6 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-10-18 12:28 ` [dpdk-dev] [PATCH v6 0/6] add Tx preparation Ananyev, Konstantin
2016-10-21 13:42 ` [dpdk-dev] [PATCH v7 " Tomasz Kulasek
2016-10-21 13:42 ` [dpdk-dev] [PATCH v7 1/6] ethdev: " Tomasz Kulasek
2016-10-21 13:42 ` [dpdk-dev] [PATCH v7 2/6] e1000: " Tomasz Kulasek
2016-10-21 13:42 ` [dpdk-dev] [PATCH v7 3/6] fm10k: " Tomasz Kulasek
2016-10-21 13:42 ` [dpdk-dev] [PATCH v7 4/6] i40e: " Tomasz Kulasek
2016-10-21 13:42 ` [dpdk-dev] [PATCH v7 5/6] ixgbe: " Tomasz Kulasek
2016-10-21 13:42 ` [dpdk-dev] [PATCH v7 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-10-21 14:46 ` [dpdk-dev] [PATCH v8 0/6] add Tx preparation Tomasz Kulasek
2016-10-21 14:46 ` [dpdk-dev] [PATCH v8 1/6] ethdev: " Tomasz Kulasek
2016-10-24 12:14 ` Ananyev, Konstantin
2016-10-24 12:49 ` Kulasek, TomaszX
2016-10-24 12:56 ` Ananyev, Konstantin
2016-10-24 14:12 ` Kulasek, TomaszX
2016-10-21 14:46 ` [dpdk-dev] [PATCH v8 2/6] e1000: " Tomasz Kulasek
2016-10-21 14:46 ` [dpdk-dev] [PATCH v8 3/6] fm10k: " Tomasz Kulasek
2016-10-21 14:46 ` [dpdk-dev] [PATCH v8 4/6] i40e: " Tomasz Kulasek
2016-10-21 14:46 ` [dpdk-dev] [PATCH v8 5/6] ixgbe: " Tomasz Kulasek
2016-10-21 14:46 ` [dpdk-dev] [PATCH v8 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-10-24 14:05 ` [dpdk-dev] [PATCH v9 0/6] add Tx preparation Tomasz Kulasek
2016-10-24 14:05 ` [dpdk-dev] [PATCH v9 1/6] ethdev: " Tomasz Kulasek
2016-10-24 14:05 ` [dpdk-dev] [PATCH v9 2/6] e1000: " Tomasz Kulasek
2016-10-24 14:05 ` [dpdk-dev] [PATCH v9 3/6] fm10k: " Tomasz Kulasek
2016-10-24 14:05 ` [dpdk-dev] [PATCH v9 4/6] i40e: " Tomasz Kulasek
2016-10-24 14:05 ` [dpdk-dev] [PATCH v9 5/6] ixgbe: " Tomasz Kulasek
2016-10-24 14:05 ` [dpdk-dev] [PATCH v9 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-10-24 16:51 ` [dpdk-dev] [PATCH v10 0/6] add Tx preparation Tomasz Kulasek
2016-10-24 16:51 ` [dpdk-dev] [PATCH v10 1/6] ethdev: " Tomasz Kulasek
2016-10-25 14:41 ` Olivier Matz
2016-10-25 17:28 ` Kulasek, TomaszX
2016-10-24 16:51 ` [dpdk-dev] [PATCH v10 2/6] e1000: " Tomasz Kulasek
2016-10-24 16:51 ` [dpdk-dev] [PATCH v10 3/6] fm10k: " Tomasz Kulasek
2016-10-24 16:51 ` [dpdk-dev] [PATCH v10 4/6] i40e: " Tomasz Kulasek
2016-10-24 16:51 ` [dpdk-dev] [PATCH v10 5/6] ixgbe: " Tomasz Kulasek
2016-10-24 16:51 ` [dpdk-dev] [PATCH v10 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-10-24 17:26 ` [dpdk-dev] [PATCH v10 0/6] add Tx preparation Ananyev, Konstantin
2016-10-26 12:56 ` [dpdk-dev] [PATCH v11 " Tomasz Kulasek
2016-10-26 12:56 ` [dpdk-dev] [PATCH v11 1/6] ethdev: " Tomasz Kulasek
2016-10-27 12:38 ` Olivier Matz
2016-10-27 15:01 ` Thomas Monjalon
2016-10-27 15:52 ` Ananyev, Konstantin
2016-10-27 16:02 ` Thomas Monjalon
2016-10-27 16:24 ` Ananyev, Konstantin
2016-10-27 16:39 ` Thomas Monjalon
2016-10-28 11:29 ` Ananyev, Konstantin
2016-10-28 11:34 ` Ananyev, Konstantin
2016-10-28 12:23 ` Thomas Monjalon
2016-10-28 12:59 ` Ananyev, Konstantin
2016-10-28 13:42 ` Thomas Monjalon
2016-11-01 12:57 ` Ananyev, Konstantin
2016-11-04 11:35 ` Thomas Monjalon
2016-10-27 16:39 ` Kulasek, TomaszX
2016-10-28 10:15 ` Ananyev, Konstantin
2016-10-28 10:22 ` Kulasek, TomaszX
2016-10-28 10:22 ` Thomas Monjalon
2016-10-28 10:28 ` Ananyev, Konstantin
2016-10-28 11:02 ` Richardson, Bruce
2016-10-28 11:14 ` Jerin Jacob
2016-10-27 16:29 ` Kulasek, TomaszX
2016-10-26 12:56 ` [dpdk-dev] [PATCH v11 2/6] e1000: " Tomasz Kulasek
2016-10-26 12:56 ` [dpdk-dev] [PATCH v11 3/6] fm10k: " Tomasz Kulasek
2016-10-26 12:56 ` [dpdk-dev] [PATCH v11 4/6] i40e: " Tomasz Kulasek
2016-10-26 12:56 ` [dpdk-dev] [PATCH v11 5/6] ixgbe: " Tomasz Kulasek
2016-10-26 12:56 ` [dpdk-dev] [PATCH v11 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-11-23 17:36 ` [dpdk-dev] [PATCH v12 0/6] add Tx preparation Tomasz Kulasek
2016-11-23 17:36 ` [dpdk-dev] [PATCH v12 1/6] ethdev: " Tomasz Kulasek
2016-11-28 10:54 ` Thomas Monjalon
2016-12-01 16:24 ` Thomas Monjalon
2016-12-01 19:20 ` Kulasek, TomaszX
2016-12-01 19:52 ` Thomas Monjalon
2016-12-01 21:56 ` Jerin Jacob
2016-12-01 22:31 ` Kulasek, TomaszX
2016-12-01 23:50 ` Thomas Monjalon
2016-12-09 13:25 ` Kulasek, TomaszX
2016-12-02 0:10 ` Ananyev, Konstantin
2016-12-22 13:14 ` Thomas Monjalon
2016-12-22 13:37 ` Jerin Jacob
2016-12-01 16:26 ` Thomas Monjalon
2016-12-01 16:28 ` Thomas Monjalon
2016-12-02 1:06 ` Ananyev, Konstantin
2016-12-02 8:24 ` Olivier Matz
2016-12-02 16:17 ` Ananyev, Konstantin
2016-12-08 17:24 ` Olivier Matz
2016-12-09 17:19 ` Kulasek, TomaszX
2016-12-12 11:51 ` Ananyev, Konstantin
2016-12-22 13:30 ` Thomas Monjalon
2016-12-22 14:11 ` Ananyev, Konstantin
2016-11-23 17:36 ` [dpdk-dev] [PATCH v12 2/6] e1000: " Tomasz Kulasek
2016-11-23 17:36 ` [dpdk-dev] [PATCH v12 3/6] fm10k: " Tomasz Kulasek
2016-11-23 17:36 ` [dpdk-dev] [PATCH v12 4/6] i40e: " Tomasz Kulasek
2016-11-23 17:36 ` [dpdk-dev] [PATCH v12 5/6] ixgbe: " Tomasz Kulasek
2016-11-23 17:36 ` [dpdk-dev] [PATCH v12 6/6] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-12-07 11:13 ` Ferruh Yigit
2016-12-07 12:00 ` Mcnamara, John
2016-12-07 12:12 ` Kulasek, TomaszX
2016-12-07 12:49 ` Ananyev, Konstantin
2016-12-07 12:00 ` Kulasek, TomaszX
2016-11-28 11:03 ` [dpdk-dev] [PATCH v12 0/6] add Tx preparation Thomas Monjalon
2016-11-30 5:48 ` John Daley (johndale)
2016-11-30 10:59 ` Ananyev, Konstantin
2016-11-30 7:40 ` Adrien Mazarguil
2016-11-30 8:50 ` Thomas Monjalon
2016-11-30 10:30 ` Kulasek, TomaszX
2016-12-01 7:19 ` Adrien Mazarguil
2016-11-30 10:54 ` Ananyev, Konstantin
2016-12-01 7:15 ` Adrien Mazarguil
2016-12-01 8:58 ` Thomas Monjalon
2016-12-01 22:03 ` Jerin Jacob
2016-12-02 1:00 ` Ananyev, Konstantin [this message]
2016-12-05 15:03 ` Adrien Mazarguil
2016-12-05 16:43 ` Ananyev, Konstantin
2016-12-05 18:10 ` Adrien Mazarguil
2016-12-06 10:56 ` Ananyev, Konstantin
2016-12-06 13:59 ` Adrien Mazarguil
2016-12-06 20:31 ` Ananyev, Konstantin
2016-12-07 10:08 ` Adrien Mazarguil
2016-11-30 16:34 ` Harish Patil
2016-11-30 17:42 ` Ananyev, Konstantin
2016-11-30 18:26 ` Thomas Monjalon
2016-11-30 21:01 ` Jerin Jacob
2016-12-01 10:50 ` Ferruh Yigit
2016-12-02 23:55 ` Yong Wang
2016-12-04 12:11 ` Ananyev, Konstantin
2016-12-06 18:25 ` Yong Wang
2016-12-07 9:57 ` Ferruh Yigit
2016-12-07 10:03 ` Ananyev, Konstantin
2016-12-07 14:31 ` Alejandro Lucero
2016-12-08 18:20 ` Yong Wang
2016-12-09 14:40 ` Jan Mędala
2016-12-12 15:02 ` Ananyev, Konstantin
2016-12-16 0:15 ` Ananyev, Konstantin
2016-12-16 13:53 ` Jan Mędala
2016-12-16 15:27 ` Ananyev, Konstantin
2016-12-16 15:37 ` Jan Mędala
2016-12-12 17:29 ` Ananyev, Konstantin
2016-11-30 18:39 ` Harish Patil
2016-11-30 19:37 ` Ajit Khaparde
2016-12-01 8:24 ` Rahul Lakkireddy
2016-12-06 15:53 ` Ferruh Yigit
2016-12-07 7:55 ` Andrew Rybchenko
2016-12-07 8:11 ` Yuanhan Liu
2016-12-07 10:13 ` Ananyev, Konstantin
2016-12-07 10:18 ` Yuanhan Liu
2016-12-07 10:22 ` Ananyev, Konstantin
2016-12-13 11:59 ` Ferruh Yigit
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 0/7] " Tomasz Kulasek
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 1/7] ethdev: " Tomasz Kulasek
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 2/7] e1000: " Tomasz Kulasek
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 3/7] fm10k: " Tomasz Kulasek
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 4/7] i40e: " Tomasz Kulasek
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 5/7] ixgbe: " Tomasz Kulasek
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 6/7] vmxnet3: " Tomasz Kulasek
2016-12-13 18:15 ` Yong Wang
2016-12-20 13:36 ` Ferruh Yigit
2016-12-22 13:10 ` Thomas Monjalon
2016-12-13 17:41 ` [dpdk-dev] [PATCH v13 7/7] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 0/8] add Tx preparation Tomasz Kulasek
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 1/8] ethdev: " Tomasz Kulasek
2016-12-22 14:24 ` Thomas Monjalon
2016-12-23 18:49 ` Kulasek, TomaszX
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 2/8] e1000: " Tomasz Kulasek
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 3/8] fm10k: " Tomasz Kulasek
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 4/8] i40e: " Tomasz Kulasek
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 5/8] ixgbe: " Tomasz Kulasek
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 6/8] vmxnet3: " Tomasz Kulasek
2016-12-22 17:59 ` Yong Wang
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 7/8] ena: " Tomasz Kulasek
2016-12-22 13:05 ` [dpdk-dev] [PATCH v14 8/8] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2016-12-22 14:28 ` Thomas Monjalon
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 0/8] add Tx preparation Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 1/8] ethdev: " Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 2/8] e1000: " Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 3/8] fm10k: " Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 4/8] i40e: " Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 5/8] ixgbe: " Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 6/8] vmxnet3: " Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 7/8] ena: " Tomasz Kulasek
2016-12-23 18:40 ` [dpdk-dev] [PATCH v15 8/8] testpmd: use Tx preparation in csum engine Tomasz Kulasek
2017-01-04 19:41 ` [dpdk-dev] [PATCH v15 0/8] add Tx preparation Thomas Monjalon
2017-01-05 15:43 ` Avi Kivity
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2601191342CEEE43887BDE71AB9772583F0E2A9A@irsmsx105.ger.corp.intel.com \
--to=konstantin.ananyev@intel.com \
--cc=adrien.mazarguil@6wind.com \
--cc=alejandro.lucero@netronome.com \
--cc=dev@dpdk.org \
--cc=harish.patil@qlogic.com \
--cc=jan@semihalf.com \
--cc=jerin.jacob@caviumnetworks.com \
--cc=johndale@cisco.com \
--cc=jpa@semihalf.com \
--cc=olivier.matz@6wind.com \
--cc=rahul.lakkireddy@chelsio.com \
--cc=rasesh.mody@qlogic.com \
--cc=stephen.hurd@broadcom.com \
--cc=thomas.monjalon@6wind.com \
--cc=tomaszx.kulasek@intel.com \
--cc=yongwang@vmware.com \
--cc=yuanhan.liu@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).