From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: "Hu, Jiayu" <jiayu.hu@intel.com>,
"Kavanagh, Mark B" <mark.b.kavanagh@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>, "Tan, Jianfeng" <jianfeng.tan@intel.com>
Subject: Re: [dpdk-dev] [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
Date: Fri, 15 Sep 2017 08:17:04 +0000 [thread overview]
Message-ID: <2601191342CEEE43887BDE71AB9772584F24B2E9@irsmsx105.ger.corp.intel.com> (raw)
In-Reply-To: <ED946F0BEFE0A141B63BABBD629A2A9B387D63B3@SHSMSX151.ccr.corp.intel.com>
> -----Original Message-----
> From: Ananyev, Konstantin
> Sent: Friday, September 15, 2017 9:16 AM
> To: Hu, Jiayu <jiayu.hu@intel.com>; Kavanagh, Mark B <mark.b.kavanagh@intel.com>
> Cc: dev@dpdk.org; Tan, Jianfeng <jianfeng.tan@intel.com>
> Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
>
> Hi Jiayu,
>
> > -----Original Message-----
> > From: Hu, Jiayu
> > Sent: Friday, September 15, 2017 8:55 AM
> > To: Ananyev, Konstantin <konstantin.ananyev@intel.com>; Kavanagh, Mark B <mark.b.kavanagh@intel.com>
> > Cc: dev@dpdk.org; Tan, Jianfeng <jianfeng.tan@intel.com>
> > Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> >
> > Hi Konstantin,
> >
> > > -----Original Message-----
> > > From: Ananyev, Konstantin
> > > Sent: Friday, September 15, 2017 2:39 AM
> > > To: Kavanagh, Mark B <mark.b.kavanagh@intel.com>; Hu, Jiayu
> > > <jiayu.hu@intel.com>
> > > Cc: dev@dpdk.org; Tan, Jianfeng <jianfeng.tan@intel.com>
> > > Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > >
> > >
> > >
> > > > -----Original Message-----
> > > > From: Kavanagh, Mark B
> > > > Sent: Thursday, September 14, 2017 4:42 PM
> > > > To: Hu, Jiayu <jiayu.hu@intel.com>; Ananyev, Konstantin
> > > <konstantin.ananyev@intel.com>
> > > > Cc: dev@dpdk.org; Tan, Jianfeng <jianfeng.tan@intel.com>
> > > > Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > > >
> > > > >From: Hu, Jiayu
> > > > >Sent: Thursday, September 14, 2017 11:01 AM
> > > > >To: Ananyev, Konstantin <konstantin.ananyev@intel.com>; Kavanagh,
> > > Mark B
> > > > ><mark.b.kavanagh@intel.com>
> > > > >Cc: dev@dpdk.org; Tan, Jianfeng <jianfeng.tan@intel.com>
> > > > >Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > > > >
> > > > >Hi Konstantin and Mark,
> > > > >
> > > > >> -----Original Message-----
> > > > >> From: Ananyev, Konstantin
> > > > >> Sent: Thursday, September 14, 2017 5:36 PM
> > > > >> To: Hu, Jiayu <jiayu.hu@intel.com>
> > > > >> Cc: dev@dpdk.org; Kavanagh, Mark B <mark.b.kavanagh@intel.com>;
> > > Tan,
> > > > >> Jianfeng <jianfeng.tan@intel.com>
> > > > >> Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > > > >>
> > > > >>
> > > > >>
> > > > >> > -----Original Message-----
> > > > >> > From: Hu, Jiayu
> > > > >> > Sent: Thursday, September 14, 2017 10:29 AM
> > > > >> > To: Ananyev, Konstantin <konstantin.ananyev@intel.com>
> > > > >> > Cc: dev@dpdk.org; Kavanagh, Mark B <mark.b.kavanagh@intel.com>;
> > > Tan,
> > > > >> Jianfeng <jianfeng.tan@intel.com>
> > > > >> > Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > > > >> >
> > > > >> > Hi Konstantin,
> > > > >> >
> > > > >> > > -----Original Message-----
> > > > >> > > From: Ananyev, Konstantin
> > > > >> > > Sent: Thursday, September 14, 2017 4:47 PM
> > > > >> > > To: Hu, Jiayu <jiayu.hu@intel.com>
> > > > >> > > Cc: dev@dpdk.org; Kavanagh, Mark B
> > > <mark.b.kavanagh@intel.com>;
> > > > >> Tan,
> > > > >> > > Jianfeng <jianfeng.tan@intel.com>
> > > > >> > > Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > > > >> > >
> > > > >> > > Hi Jiayu,
> > > > >> > >
> > > > >> > > > -----Original Message-----
> > > > >> > > > From: Hu, Jiayu
> > > > >> > > > Sent: Thursday, September 14, 2017 7:07 AM
> > > > >> > > > To: Ananyev, Konstantin <konstantin.ananyev@intel.com>
> > > > >> > > > Cc: dev@dpdk.org; Kavanagh, Mark B
> > > <mark.b.kavanagh@intel.com>;
> > > > >> Tan,
> > > > >> > > Jianfeng <jianfeng.tan@intel.com>
> > > > >> > > > Subject: Re: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > > > >> > > >
> > > > >> > > > Hi Konstantin,
> > > > >> > > >
> > > > >> > > > On Thu, Sep 14, 2017 at 06:10:37AM +0800, Ananyev, Konstantin
> > > wrote:
> > > > >> > > > >
> > > > >> > > > > Hi Jiayu,
> > > > >> > > > >
> > > > >> > > > > > >
> > > > >> > > > > > >
> > > > >> > > > > > > > -----Original Message-----
> > > > >> > > > > > > > From: Ananyev, Konstantin
> > > > >> > > > > > > > Sent: Tuesday, September 12, 2017 12:18 PM
> > > > >> > > > > > > > To: Hu, Jiayu <jiayu.hu@intel.com>; dev@dpdk.org
> > > > >> > > > > > > > Cc: Kavanagh, Mark B <mark.b.kavanagh@intel.com>; Tan,
> > > > >> Jianfeng
> > > > >> > > <jianfeng.tan@intel.com>
> > > > >> > > > > > > > Subject: RE: [PATCH v3 2/5] gso: add TCP/IPv4 GSO support
> > > > >> > > > > > > >
> > > > >> > > > > > > > > result, when all of its GSOed segments are freed, the
> > > packet
> > > > >is
> > > > >> > > freed
> > > > >> > > > > > > > > automatically.
> > > > >> > > > > > > > > diff --git a/lib/librte_gso/rte_gso.c
> > > > >b/lib/librte_gso/rte_gso.c
> > > > >> > > > > > > > > index dda50ee..95f6ea6 100644
> > > > >> > > > > > > > > --- a/lib/librte_gso/rte_gso.c
> > > > >> > > > > > > > > +++ b/lib/librte_gso/rte_gso.c
> > > > >> > > > > > > > > @@ -33,18 +33,53 @@
> > > > >> > > > > > > > >
> > > > >> > > > > > > > > #include <errno.h>
> > > > >> > > > > > > > >
> > > > >> > > > > > > > > +#include <rte_log.h>
> > > > >> > > > > > > > > +
> > > > >> > > > > > > > > #include "rte_gso.h"
> > > > >> > > > > > > > > +#include "gso_common.h"
> > > > >> > > > > > > > > +#include "gso_tcp4.h"
> > > > >> > > > > > > > >
> > > > >> > > > > > > > > int
> > > > >> > > > > > > > > rte_gso_segment(struct rte_mbuf *pkt,
> > > > >> > > > > > > > > - struct rte_gso_ctx gso_ctx __rte_unused,
> > > > >> > > > > > > > > + struct rte_gso_ctx gso_ctx,
> > > > >> > > > > > > > > struct rte_mbuf **pkts_out,
> > > > >> > > > > > > > > uint16_t nb_pkts_out)
> > > > >> > > > > > > > > {
> > > > >> > > > > > > > > + struct rte_mempool *direct_pool, *indirect_pool;
> > > > >> > > > > > > > > + struct rte_mbuf *pkt_seg;
> > > > >> > > > > > > > > + uint16_t gso_size;
> > > > >> > > > > > > > > + uint8_t ipid_delta;
> > > > >> > > > > > > > > + int ret = 1;
> > > > >> > > > > > > > > +
> > > > >> > > > > > > > > if (pkt == NULL || pkts_out == NULL || nb_pkts_out
> > > < 1)
> > > > >> > > > > > > > > return -EINVAL;
> > > > >> > > > > > > > >
> > > > >> > > > > > > > > - pkts_out[0] = pkt;
> > > > >> > > > > > > > > + if (gso_ctx.gso_size >= pkt->pkt_len ||
> > > > >> > > > > > > > > + (pkt->packet_type &
> > > gso_ctx.gso_types) !=
> > > > >> > > > > > > > > + pkt->packet_type) {
> > > > >> > > > > > > > > + pkts_out[0] = pkt;
> > > > >> > > > > > > > > + return ret;
> > > > >> > > > > > > > > + }
> > > > >> > > > > > > > > +
> > > > >> > > > > > > > > + direct_pool = gso_ctx.direct_pool;
> > > > >> > > > > > > > > + indirect_pool = gso_ctx.indirect_pool;
> > > > >> > > > > > > > > + gso_size = gso_ctx.gso_size;
> > > > >> > > > > > > > > + ipid_delta = gso_ctx.ipid_flag ==
> > > RTE_GSO_IPID_INCREASE;
> > > > >> > > > > > > > > +
> > > > >> > > > > > > > > + if (is_ipv4_tcp(pkt->packet_type)) {
> > > > >> > > > > > > >
> > > > >> > > > > > > > Probably we need here:
> > > > >> > > > > > > > If (is_ipv4_tcp(pkt->packet_type) && (gso_ctx->gso_types
> > > &
> > > > >> > > DEV_TX_OFFLOAD_TCP_TSO) != 0) {...
> > > > >> > > > > > >
> > > > >> > > > > > > Sorry, actually it probably should be:
> > > > >> > > > > > > If (pkt->ol_flags & (PKT_TX_TCP_SEG | PKT_TX_IPV4) ==
> > > > >> PKT_TX_IPV4
> > > > >> > > &&
> > > > >> > > > > > > (gso_ctx->gso_types & DEV_TX_OFFLOAD_TCP_TSO) != 0)
> > > {...
> > > > >> > > > > >
> > > > >> > > > > > I don't quite understand why the GSO library should be aware if
> > > > >the
> > > > >> TSO
> > > > >> > > > > > flag is set or not. Applications can query device TSO capability
> > > > >> before
> > > > >> > > > > > they call the GSO library. Do I misundertsand anything?
> > > > >> > > > > >
> > > > >> > > > > > Additionally, we don't need to check if the packet is a TCP/IPv4
> > > > >> packet
> > > > >> > > here?
> > > > >> > > > >
> > > > >> > > > > Well, right now PMD we doesn't rely on ptype to figure out what
> > > > >type
> > > > >> of
> > > > >> > > packet and
> > > > >> > > > > what TX offload have to be performed.
> > > > >> > > > > Instead it looks at TX part of ol_flags, and
> > > > >> > > > > My thought was that as what we doing is actually TSO in SW, it
> > > would
> > > > >> be
> > > > >> > > good
> > > > >> > > > > to use the same API here too.
> > > > >> > > > > Also with that approach, by setting ol_flags properly user can use
> > > > >the
> > > > >> > > same gso_ctx and still
> > > > >> > > > > specify what segmentation to perform on a per-packet basis.
> > > > >> > > > >
> > > > >> > > > > Alternative way is to rely on ptype to distinguish should
> > > > >segmentation
> > > > >> be
> > > > >> > > performed on that package or not.
> > > > >> > > > > The only advantage I see here is that if someone would like to
> > > add
> > > > >> GSO
> > > > >> > > for some new protocol,
> > > > >> > > > > he wouldn't need to introduce new TX flag value for
> > > mbuf.ol_flags.
> > > > >> > > > > Though he still would need to update TX_OFFLOAD_* capabilities
> > > and
> > > > >> > > probably packet_type definitions.
> > > > >> > > > >
> > > > >> > > > > So from my perspective first variant (use HW TSO API) is more
> > > > >> plausible.
> > > > >> > > > > Wonder what is your and Mark opinions here?
> > > > >> > > >
> > > > >> > > > In the first choice, you mean:
> > > > >> > > > the GSO library uses gso_ctx->gso_types and mbuf->ol_flags to call
> > > a
> > > > >> > > specific GSO
> > > > >> > > > segmentation function (e.g. gso_tcp4_segment(), gso_tunnel_xxx())
> > > for
> > > > >> > > each input packet.
> > > > >> > > > Applications should parse the packet type, and set an exactly
> > > correct
> > > > >> > > DEV_TX_OFFLOAD_*_TSO
> > > > >> > > > flag to gso_types and ol_flags according to the packet type. That is,
> > > > >the
> > > > >> > > value of gso_types
> > > > >> > > > is on a per-packet basis. Using gso_ctx->gso_types and mbuf-
> > > >ol_flags
> > > > >> at
> > > > >> > > the same time
> > > > >> > > > is because that DEV_TX_OFFLOAD_*_TSO only tells tunnelling type
> > > and
> > > > >> the
> > > > >> > > inner L4 type, and
> > > > >> > > > we need to know L3 type by ol_flags. With this design, HW
> > > > >> segmentation
> > > > >> > > and SW segmentation
> > > > >> > > > are indeed consistent.
> > > > >> > > >
> > > > >> > > > If I understand it correctly, applications need to set 'ol_flags =
> > > > >> > > PKT_TX_IPV4' and
> > > > >> > > > 'gso_types = DEV_TX_OFFLOAD_VXLAN_TNL_TSO' for a
> > > > >> > > "ether+ipv4+udp+vxlan+ether+ipv4+
> > > > >> > > > tcp+payload" packet. But PKT_TX_IPV4 just present the inner L3
> > > type
> > > > >for
> > > > >> > > tunneled packet.
> > > > >> > > > How about the outer L3 type? Always assume the inner and the
> > > outer L3
> > > > >> > > type are the same?
> > > > >> > >
> > > > >> > > It think that for that case you'll have to set in ol_flags:
> > > > >> > >
> > > > >> > > PKT_TX_IPV4 | PKT_TX_OUTER_IPV4 | PKT_TX_TUNNEL_VXLAN |
> > > > >> > > PKT_TX_TCP_SEG
> > > > >> >
> > > > >> > OK, so it means PKT_TX_TCP_SEG is also used for tunneled TSO. The
> > > > >> > GSO library doesn't need gso_types anymore.
> > > > >>
> > > > >> You still might need gso_ctx.gso_types to let user limit what types of
> > > > >> segmentation
> > > > >> that particular gso_ctx supports.
> > > > >> An alternative would be to assume that each gso_ctx supports all
> > > > >> currently implemented segmentations.
> > > > >> This is possible too, but probably not very convenient to the user.
> > > > >
> > > > >Hmm, make sense.
> > > > >
> > > > >One thing to confirm: the value of gso_types should be
> > > DEV_TX_OFFLOAD_*_TSO,
> > > > >or new macros?
> > > >
> > > > Hi Jiayu, Konstantin,
> > > >
> > > > I think that the existing macros are fine, as they provide a consistent view
> > > of segmentation capabilities to the application/user.
> > >
> > > +1
> > > I also think it is better to re-use DEV_TX_OFFLOAD_*_TSO.
> >
> > There might be an 'issue', if we use 'PKT_TX_TCP_SEG' to tell the
> > GSO library to segment a packet or not. Given the scenario that
> > an application only wants to do GSO and doesn't want to use TSO.
> > The application sets 'mbuf->ol_flags=PKT_TX_TCP_SEG' and doesn't
> > set mbuf->tso_segsz. Then the GSO library segments the packet, and
> > all output GSO segments have the same ol_flags as the input packet
> > (in current GSO library design). Then the output GSO segments are
> > transmitted to rte_eth_tx_prepare(). If the NIC is i40e, its TX prepare function,
> > i40e_prep_pkts, checks if mbuf->tso_segsz is in the range of I40E_MIN_TSO_MSS
> > and I40E_MAX_TSO_MSS, when PKT_TX_TCP_SEG is set. So an error happens in
> > this scenario, since tso_segsz is 0.
> >
> > In fact, it may confuse the PMD driver when set PKT_TX_TCP_SEG but don't want
> > to do TSO. One solution is that the GSO library removes the PKT_TX_TCP_SEG flag
> > for all GSO segments after finishes segmenting.
>
> Yes, that was my thought too: after successful segmentation we probably
> need to cleanup related ol_flags.
In fact, we just don't need to set these flags in our newly created segments.
> Konstantin
>
> > Wonder you and Mark's opinion.
> >
> > Thanks,
> > Jiayu
> > >
> > > >
> > > > I was initially concerned that they might be too coarse-grained (i.e. only
> > > IPv4 is currently supported, and not IPv6), but as per Konstantin's
> > > > previous example, the DEV_TX_OFFLOAD_*_TSO macros can be used in
> > > concert with the packet type to determine whether a packet should
> > > > be fragmented or not.
> > > >
> > > > Thanks,
> > > > Mark
> > > >
> > > > >
> > > > >Jiayu
> > > > >> Konstantin
> > > > >>
> > > > >> >
> > > > >> > The first choice makes HW and SW segmentation are totally the same.
> > > > >> > Applications just need to parse the packet and set proper ol_flags, and
> > > > >> > the GSO library uses ol_flags to decide which segmentation function to
> > > > >use.
> > > > >> > I think it's better than the second choice which depending on ptype to
> > > > >> > choose segmentation function.
> > > > >> >
> > > > >> > Jiayu
> > > > >> > >
> > > > >> > > Konstantin
> > > > >> > >
> > > > >> > > >
> > > > >> > > > Jiayu
> > > > >> > > > > Konstantin
next prev parent reply other threads:[~2017-09-15 8:17 UTC|newest]
Thread overview: 157+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-24 14:15 [dpdk-dev] [PATCH 0/5] Support TCP/IPv4, VxLAN and GRE GSO in DPDK Jiayu Hu
2017-08-24 14:15 ` [dpdk-dev] [PATCH 1/5] lib: add Generic Segmentation Offload API framework Jiayu Hu
2017-08-30 1:38 ` Ananyev, Konstantin
2017-08-30 7:57 ` Jiayu Hu
2017-08-24 14:15 ` [dpdk-dev] [PATCH 2/5] gso/lib: add TCP/IPv4 GSO support Jiayu Hu
2017-08-30 1:38 ` Ananyev, Konstantin
2017-08-30 2:55 ` Jiayu Hu
2017-08-30 9:25 ` Kavanagh, Mark B
2017-08-30 9:39 ` Ananyev, Konstantin
2017-08-30 9:59 ` Ananyev, Konstantin
2017-08-30 13:27 ` Kavanagh, Mark B
2017-08-30 9:03 ` Jiayu Hu
2017-09-04 3:31 ` Jiayu Hu
2017-09-04 9:54 ` Ananyev, Konstantin
2017-09-05 1:09 ` Hu, Jiayu
2017-09-11 13:04 ` Ananyev, Konstantin
2017-08-24 14:15 ` [dpdk-dev] [PATCH 3/5] lib/gso: add VxLAN " Jiayu Hu
2017-08-24 14:15 ` [dpdk-dev] [PATCH 4/5] lib/gso: add GRE " Jiayu Hu
2017-08-24 14:15 ` [dpdk-dev] [PATCH 5/5] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Jiayu Hu
2017-08-30 1:37 ` [dpdk-dev] [PATCH 0/5] Support TCP/IPv4, VxLAN and GRE GSO in DPDK Ananyev, Konstantin
2017-08-30 7:36 ` Jiayu Hu
2017-08-30 10:49 ` Ananyev, Konstantin
2017-08-30 13:32 ` Kavanagh, Mark B
2017-09-05 7:57 ` [dpdk-dev] [PATCH v2 " Jiayu Hu
2017-09-05 7:57 ` [dpdk-dev] [PATCH v2 1/5] gso: add Generic Segmentation Offload API framework Jiayu Hu
2017-09-05 7:57 ` [dpdk-dev] [PATCH v2 2/5] gso: add TCP/IPv4 GSO support Jiayu Hu
2017-09-05 7:57 ` [dpdk-dev] [PATCH v2 3/5] gso: add VxLAN " Jiayu Hu
2017-09-05 7:57 ` [dpdk-dev] [PATCH v2 4/5] gso: add GRE " Jiayu Hu
2017-09-05 7:57 ` [dpdk-dev] [PATCH v2 5/5] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Jiayu Hu
2017-09-12 2:43 ` [dpdk-dev] [PATCH v3 0/5] Support TCP/IPv4, VxLAN and GRE GSO in DPDK Jiayu Hu
2017-09-12 2:43 ` [dpdk-dev] [PATCH v3 1/5] gso: add Generic Segmentation Offload API framework Jiayu Hu
2017-09-12 10:36 ` Ananyev, Konstantin
2017-09-13 2:11 ` Jiayu Hu
2017-09-14 18:33 ` Ferruh Yigit
2017-09-15 1:12 ` Hu, Jiayu
2017-09-12 2:43 ` [dpdk-dev] [PATCH v3 2/5] gso: add TCP/IPv4 GSO support Jiayu Hu
2017-09-12 11:17 ` Ananyev, Konstantin
2017-09-13 2:48 ` Jiayu Hu
2017-09-13 9:38 ` Ananyev, Konstantin
2017-09-13 10:23 ` Hu, Jiayu
2017-09-13 14:52 ` Kavanagh, Mark B
2017-09-13 15:13 ` Ananyev, Konstantin
2017-09-14 0:59 ` Hu, Jiayu
2017-09-14 8:35 ` Kavanagh, Mark B
2017-09-14 8:39 ` Ananyev, Konstantin
2017-09-14 9:00 ` Kavanagh, Mark B
2017-09-14 9:10 ` Ananyev, Konstantin
2017-09-14 9:35 ` Kavanagh, Mark B
2017-09-12 14:17 ` Ananyev, Konstantin
2017-09-13 10:44 ` Jiayu Hu
2017-09-13 22:10 ` Ananyev, Konstantin
2017-09-14 6:07 ` Jiayu Hu
2017-09-14 8:47 ` Ananyev, Konstantin
2017-09-14 9:29 ` Hu, Jiayu
2017-09-14 9:35 ` Ananyev, Konstantin
2017-09-14 10:01 ` Hu, Jiayu
2017-09-14 15:42 ` Kavanagh, Mark B
2017-09-14 18:38 ` Ananyev, Konstantin
2017-09-15 7:54 ` Hu, Jiayu
2017-09-15 8:15 ` Ananyev, Konstantin
2017-09-15 8:17 ` Ananyev, Konstantin [this message]
2017-09-15 8:38 ` Hu, Jiayu
2017-09-14 8:51 ` Kavanagh, Mark B
2017-09-14 9:45 ` Hu, Jiayu
2017-09-12 2:43 ` [dpdk-dev] [PATCH v3 3/5] gso: add VxLAN " Jiayu Hu
2017-09-12 2:43 ` [dpdk-dev] [PATCH v3 4/5] gso: add GRE " Jiayu Hu
2017-09-12 2:43 ` [dpdk-dev] [PATCH v3 5/5] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Jiayu Hu
2017-09-14 18:33 ` Ferruh Yigit
2017-09-15 1:13 ` Hu, Jiayu
2017-09-19 7:32 ` [dpdk-dev] [PATCH v4 0/5] Support TCP/IPv4, VxLAN and GRE GSO in DPDK Jiayu Hu
2017-09-19 7:32 ` [dpdk-dev] [PATCH v4 1/5] gso: add Generic Segmentation Offload API framework Jiayu Hu
2017-09-19 7:32 ` [dpdk-dev] [PATCH v4 2/5] gso: add TCP/IPv4 GSO support Jiayu Hu
2017-09-20 7:03 ` Yao, Lei A
2017-09-19 7:32 ` [dpdk-dev] [PATCH v4 3/5] gso: add VxLAN " Jiayu Hu
2017-09-20 3:11 ` Tan, Jianfeng
2017-09-20 3:17 ` Hu, Jiayu
2017-09-19 7:32 ` [dpdk-dev] [PATCH v4 4/5] gso: add GRE " Jiayu Hu
2017-09-20 2:53 ` Tan, Jianfeng
2017-09-20 6:01 ` Hu, Jiayu
2017-09-19 7:32 ` [dpdk-dev] [PATCH v4 5/5] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Jiayu Hu
2017-09-28 22:13 ` [dpdk-dev] [PATCH v5 0/6] Support TCP/IPv4, VxLAN and GRE GSO in DPDK Mark Kavanagh
2017-10-02 16:45 ` [dpdk-dev] [PATCH v6 0/6] Support TCP/IPv4, VxLAN, " Mark Kavanagh
2017-10-02 16:45 ` [dpdk-dev] [PATCH v6 1/6] gso: add Generic Segmentation Offload API framework Mark Kavanagh
2017-10-04 13:11 ` Ananyev, Konstantin
2017-10-04 13:21 ` Kavanagh, Mark B
2017-10-02 16:45 ` [dpdk-dev] [PATCH v6 2/6] gso: add TCP/IPv4 GSO support Mark Kavanagh
2017-10-04 13:32 ` Ananyev, Konstantin
2017-10-04 14:30 ` Kavanagh, Mark B
2017-10-04 14:49 ` Ananyev, Konstantin
2017-10-04 14:59 ` Kavanagh, Mark B
2017-10-04 13:35 ` Ananyev, Konstantin
2017-10-04 14:22 ` Kavanagh, Mark B
2017-10-02 16:45 ` [dpdk-dev] [PATCH v6 3/6] gso: add VxLAN " Mark Kavanagh
2017-10-04 14:12 ` Ananyev, Konstantin
2017-10-04 14:35 ` Kavanagh, Mark B
2017-10-04 16:13 ` Kavanagh, Mark B
2017-10-04 16:17 ` Ananyev, Konstantin
2017-10-02 16:45 ` [dpdk-dev] [PATCH v6 4/6] gso: add GRE " Mark Kavanagh
2017-10-04 14:15 ` Ananyev, Konstantin
2017-10-04 14:36 ` Kavanagh, Mark B
2017-10-02 16:45 ` [dpdk-dev] [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Mark Kavanagh
2017-10-04 15:08 ` Ananyev, Konstantin
2017-10-04 16:23 ` Kavanagh, Mark B
2017-10-04 16:26 ` Ananyev, Konstantin
2017-10-04 16:51 ` Kavanagh, Mark B
2017-10-02 16:45 ` [dpdk-dev] [PATCH v6 6/6] doc: add GSO programmer's guide Mark Kavanagh
2017-10-04 13:51 ` Mcnamara, John
2017-10-05 11:02 ` [dpdk-dev] [PATCH v7 0/6] Support TCP/IPv4, VxLAN, and GRE GSO in DPDK Mark Kavanagh
2017-10-05 11:02 ` [dpdk-dev] [PATCH v7 1/6] gso: add Generic Segmentation Offload API framework Mark Kavanagh
2017-10-05 11:02 ` [dpdk-dev] [PATCH v7 2/6] gso: add TCP/IPv4 GSO support Mark Kavanagh
2017-10-05 11:02 ` [dpdk-dev] [PATCH v7 3/6] gso: add VxLAN " Mark Kavanagh
2017-10-05 11:02 ` [dpdk-dev] [PATCH v7 4/6] gso: add GRE " Mark Kavanagh
2017-10-05 11:02 ` [dpdk-dev] [PATCH v7 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Mark Kavanagh
2017-10-05 11:02 ` [dpdk-dev] [PATCH v7 6/6] doc: add GSO programmer's guide Mark Kavanagh
2017-10-05 13:22 ` [dpdk-dev] [PATCH v7 0/6] Support TCP/IPv4, VxLAN, and GRE GSO in DPDK Ananyev, Konstantin
2017-10-05 14:39 ` Kavanagh, Mark B
2017-10-05 15:43 ` [dpdk-dev] [PATCH v8 " Mark Kavanagh
2017-10-05 17:12 ` Ananyev, Konstantin
2017-10-05 20:16 ` Kavanagh, Mark B
2017-10-05 20:36 ` [dpdk-dev] [PATCH v9 " Mark Kavanagh
2017-10-05 22:24 ` Ananyev, Konstantin
2017-10-06 8:24 ` [dpdk-dev] FW: " Kavanagh, Mark B
2017-10-06 10:35 ` [dpdk-dev] " Kavanagh, Mark B
2017-10-06 23:32 ` Ferruh Yigit
2017-10-06 23:34 ` Ferruh Yigit
2017-10-07 14:56 ` [dpdk-dev] [PATCH v10 " Jiayu Hu
2017-10-07 14:56 ` [dpdk-dev] [PATCH v10 1/6] gso: add Generic Segmentation Offload API framework Jiayu Hu
2017-10-07 14:56 ` [dpdk-dev] [PATCH v10 2/6] gso: add TCP/IPv4 GSO support Jiayu Hu
2017-10-07 14:56 ` [dpdk-dev] [PATCH v10 3/6] gso: add VxLAN " Jiayu Hu
2017-10-07 14:56 ` [dpdk-dev] [PATCH v10 4/6] gso: add GRE " Jiayu Hu
2017-10-07 14:56 ` [dpdk-dev] [PATCH v10 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Jiayu Hu
2017-10-07 14:56 ` [dpdk-dev] [PATCH v10 6/6] doc: add GSO programmer's guide Jiayu Hu
2017-10-08 3:40 ` [dpdk-dev] [PATCH v10 0/6] Support TCP/IPv4, VxLAN, and GRE GSO in DPDK Ferruh Yigit
2017-10-05 20:36 ` [dpdk-dev] [PATCH v9 1/6] gso: add Generic Segmentation Offload API framework Mark Kavanagh
2017-10-05 20:36 ` [dpdk-dev] [PATCH v9 2/6] gso: add TCP/IPv4 GSO support Mark Kavanagh
2017-10-05 20:36 ` [dpdk-dev] [PATCH v9 3/6] gso: add VxLAN " Mark Kavanagh
2017-10-05 20:36 ` [dpdk-dev] [PATCH v9 4/6] gso: add GRE " Mark Kavanagh
2017-10-05 20:36 ` [dpdk-dev] [PATCH v9 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Mark Kavanagh
2017-10-05 20:36 ` [dpdk-dev] [PATCH v9 6/6] doc: add GSO programmer's guide Mark Kavanagh
2017-10-06 13:34 ` Mcnamara, John
2017-10-06 13:41 ` Kavanagh, Mark B
2017-10-05 15:43 ` [dpdk-dev] [PATCH v8 1/6] gso: add Generic Segmentation Offload API framework Mark Kavanagh
2017-10-05 15:44 ` [dpdk-dev] [PATCH v8 2/6] gso: add TCP/IPv4 GSO support Mark Kavanagh
2017-10-05 15:44 ` [dpdk-dev] [PATCH v8 3/6] gso: add VxLAN " Mark Kavanagh
2017-10-05 15:44 ` [dpdk-dev] [PATCH v8 4/6] gso: add GRE " Mark Kavanagh
2017-10-05 15:44 ` [dpdk-dev] [PATCH v8 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Mark Kavanagh
2017-10-05 15:44 ` [dpdk-dev] [PATCH v8 6/6] doc: add GSO programmer's guide Mark Kavanagh
2017-10-05 17:57 ` Mcnamara, John
2017-09-28 22:13 ` [dpdk-dev] [PATCH v5 1/6] gso: add Generic Segmentation Offload API framework Mark Kavanagh
2017-09-28 22:13 ` [dpdk-dev] [PATCH v5 2/6] gso: add TCP/IPv4 GSO support Mark Kavanagh
2017-09-29 3:12 ` Jiayu Hu
2017-09-29 9:05 ` Kavanagh, Mark B
2017-09-28 22:13 ` [dpdk-dev] [PATCH v5 3/6] gso: add VxLAN " Mark Kavanagh
2017-09-28 22:13 ` [dpdk-dev] [PATCH v5 4/6] gso: add GRE " Mark Kavanagh
2017-09-28 22:13 ` [dpdk-dev] [PATCH v5 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO Mark Kavanagh
2017-09-28 22:13 ` [dpdk-dev] [PATCH v5 6/6] doc: add GSO programmer's guide Mark Kavanagh
2017-09-28 22:18 ` [dpdk-dev] [PATCH v5 2/6] gso: add TCP/IPv4 GSO support Mark Kavanagh
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=2601191342CEEE43887BDE71AB9772584F24B2E9@irsmsx105.ger.corp.intel.com \
--to=konstantin.ananyev@intel.com \
--cc=dev@dpdk.org \
--cc=jianfeng.tan@intel.com \
--cc=jiayu.hu@intel.com \
--cc=mark.b.kavanagh@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).