From: "Zhang, Qi Z" <qi.z.zhang@intel.com>
To: "Zhang, AlvinX" <alvinx.zhang@intel.com>, "Guo, Jia" <jia.guo@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v5] net/iavf: support outer IP hash for no inner GTPU
Date: Mon, 14 Sep 2020 10:33:02 +0000 [thread overview]
Message-ID: <66caca1e210445e38955d8cc5cdb3319@intel.com> (raw)
In-Reply-To: <20200914090316.49740-1-alvinx.zhang@intel.com>
> -----Original Message-----
> From: Zhang, AlvinX <alvinx.zhang@intel.com>
> Sent: Monday, September 14, 2020 5:03 PM
> To: Guo, Jia <jia.guo@intel.com>; Zhang, Qi Z <qi.z.zhang@intel.com>
> Cc: dev@dpdk.org; Zhang, AlvinX <alvinx.zhang@intel.com>
> Subject: [PATCH v5] net/iavf: support outer IP hash for no inner GTPU
>
> From: Alvin Zhang <alvinx.zhang@intel.com>
>
> Outer IP hash can be configured as input sets for no inner GTPU packets.
>
> Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
> ---
>
> V2: Modify codes according to comments.
> V3: Refact all codes.
> V4: Fix compatibility issues.
> V5: Modify codes according to comments.
>
> drivers/net/iavf/iavf_hash.c | 48 ++++++++++++++++++++++++++------------------
> 1 file changed, 28 insertions(+), 20 deletions(-)
>
> diff --git a/drivers/net/iavf/iavf_hash.c b/drivers/net/iavf/iavf_hash.c index
> ff77d71..d0a6396 100644
> --- a/drivers/net/iavf/iavf_hash.c
> +++ b/drivers/net/iavf/iavf_hash.c
> @@ -348,6 +348,7 @@ struct virtchnl_proto_hdrs ipv6_pfcp_tmplt = {
> {iavf_pattern_eth_vlan_ipv4_udp, IAVF_RSS_TYPE_VLAN_IPV4_UDP,
> &outer_ipv4_udp_tmplt},
> {iavf_pattern_eth_vlan_ipv4_tcp, IAVF_RSS_TYPE_VLAN_IPV4_TCP,
> &outer_ipv4_tcp_tmplt},
> {iavf_pattern_eth_vlan_ipv4_sctp, IAVF_RSS_TYPE_VLAN_IPV4_SCTP,
> &outer_ipv4_sctp_tmplt},
> + {iavf_pattern_eth_ipv4_gtpu, ETH_RSS_IPV4,
> &outer_ipv4_udp_tmplt},
> {iavf_pattern_eth_ipv4_gtpu_ipv4, IAVF_RSS_TYPE_GTPU_IPV4,
> &inner_ipv4_tmplt},
> {iavf_pattern_eth_ipv4_gtpu_ipv4_udp,
> IAVF_RSS_TYPE_GTPU_IPV4_UDP, &inner_ipv4_udp_tmplt},
> {iavf_pattern_eth_ipv4_gtpu_ipv4_tcp,
> IAVF_RSS_TYPE_GTPU_IPV4_TCP, &inner_ipv4_tcp_tmplt},
> @@ -374,6 +375,7 @@ struct virtchnl_proto_hdrs ipv6_pfcp_tmplt = {
> {iavf_pattern_eth_vlan_ipv6_udp, IAVF_RSS_TYPE_VLAN_IPV6_UDP,
> &outer_ipv6_udp_tmplt},
> {iavf_pattern_eth_vlan_ipv6_tcp, IAVF_RSS_TYPE_VLAN_IPV6_TCP,
> &outer_ipv6_tcp_tmplt},
> {iavf_pattern_eth_vlan_ipv6_sctp, IAVF_RSS_TYPE_VLAN_IPV6_SCTP,
> &outer_ipv6_sctp_tmplt},
> + {iavf_pattern_eth_ipv6_gtpu, ETH_RSS_IPV6,
> &outer_ipv6_udp_tmplt},
> {iavf_pattern_eth_ipv4_gtpu_ipv6, IAVF_RSS_TYPE_GTPU_IPV6,
> &inner_ipv6_tmplt},
> {iavf_pattern_eth_ipv4_gtpu_ipv6_udp,
> IAVF_RSS_TYPE_GTPU_IPV6_UDP, &inner_ipv6_udp_tmplt},
> {iavf_pattern_eth_ipv4_gtpu_ipv6_tcp,
> IAVF_RSS_TYPE_GTPU_IPV6_TCP, &inner_ipv6_tcp_tmplt},
> @@ -698,31 +700,37 @@ struct virtchnl_proto_hdrs
> *iavf_hash_default_hdrs[] = {
> struct virtchnl_proto_hdr *hdr2;
> int i;
>
> - if (!(phint & IAVF_PHINT_GTPU_MSK) ||
> - proto_hdrs->tunnel_level == 0)
> + if (!(phint & IAVF_PHINT_GTPU_MSK))
> return;
>
> - /* shift headers 1 layer */
> - for (i = proto_hdrs->count; i > 0; i--) {
> - hdr1 = &proto_hdrs->proto_hdr[i];
> - hdr2 = &proto_hdrs->proto_hdr[i - 1];
> + if (proto_hdrs->tunnel_level == TUNNEL_LEVEL_INNER) {
> + /* shift headers 1 layer */
> + for (i = proto_hdrs->count; i > 0; i--) {
> + hdr1 = &proto_hdrs->proto_hdr[i];
> + hdr2 = &proto_hdrs->proto_hdr[i - 1];
>
> - *hdr1 = *hdr2;
> - }
> + *hdr1 = *hdr2;
> + }
>
> - /* adding gtpu header at layer 0 */
> - proto_hdrs->count++;
> - hdr1 = &proto_hdrs->proto_hdr[0];
> - hdr1->field_selector = 0;
> -
> - if (phint & IAVF_PHINT_GTPU_EH_DWN)
> - VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH_PDU_DWN);
> - else if (phint & IAVF_PHINT_GTPU_EH_UP)
> - VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH_PDU_UP);
> - else if (phint & IAVF_PHINT_GTPU_EH)
> - VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH);
> - else if (phint & IAVF_PHINT_GTPU)
> + /* adding gtpu header at layer 0 */
> + proto_hdrs->count++;
> + hdr1 = &proto_hdrs->proto_hdr[0];
> + hdr1->field_selector = 0;
> +
> + if (phint & IAVF_PHINT_GTPU_EH_DWN)
> + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH_PDU_DWN);
> + else if (phint & IAVF_PHINT_GTPU_EH_UP)
> + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH_PDU_UP);
> + else if (phint & IAVF_PHINT_GTPU_EH)
> + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH);
> + else if (phint & IAVF_PHINT_GTPU)
> + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_IP);
> + } else {
> + hdr1 = &proto_hdrs->proto_hdr[proto_hdrs->count];
> + hdr1->field_selector = 0;
> + proto_hdrs->count++;
> VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_IP);
Its better to also check phint for outer case, could be implemented as below
If (inner) {
Shift
Hdr = hdrs[0]
}
else {
Hdr = hdrs[count]
}
hdr-count++
if (phint & IAVF_PHINT_GTPU_EH_DWN)
VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH_PDU_DWN);
else if (phint & IAVF_PHINT_GTPU_EH_UP)
VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH_PDU_UP);
else if (phint & IAVF_PHINT_GTPU_EH)
VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_EH);
else if (phint & IAVF_PHINT_GTPU)
VIRTCHNL_SET_PROTO_HDR_TYPE(hdr1, GTPU_IP);
> + }
> }
>
> static void iavf_refine_proto_hdrs(struct virtchnl_proto_hdrs *proto_hdrs,
> --
> 1.8.3.1
next prev parent reply other threads:[~2020-09-14 10:33 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-08 9:50 [dpdk-dev] [PATCH v3 1/3] common/iavf: add GTPC support alvinx.zhang
2020-09-08 9:50 ` [dpdk-dev] [PATCH v3 2/3] net/iavf: support outer IP hash for GTPC alvinx.zhang
2020-09-08 9:50 ` [dpdk-dev] [PATCH v3 3/3] net/iavf: support outer IP hash for GTPU alvinx.zhang
2020-09-10 3:26 ` [dpdk-dev] [PATCH v4] " alvinx.zhang
2020-09-10 5:24 ` Guo, Jia
2020-09-10 9:24 ` Zhang, AlvinX
2020-09-14 9:03 ` [dpdk-dev] [PATCH v5] net/iavf: support outer IP hash for no inner GTPU alvinx.zhang
2020-09-14 10:33 ` Zhang, Qi Z [this message]
2020-09-15 1:54 ` Zhang, AlvinX
2020-09-15 2:43 ` [dpdk-dev] [PATCH v6] " alvinx.zhang
2020-09-15 2:58 ` Zhang, Qi Z
2020-09-10 3:36 ` [dpdk-dev] [PATCH v4 1/2] common/iavf: add GTPC support alvinx.zhang
2020-09-10 3:36 ` [dpdk-dev] [PATCH v4 2/2] net/iavf: support outer IP hash for GTPC alvinx.zhang
2020-09-10 5:14 ` Guo, Jia
2020-09-10 9:29 ` Zhang, AlvinX
2020-09-14 9:12 ` [dpdk-dev] [PATCH v5] " alvinx.zhang
2020-09-14 10:48 ` Zhang, Qi Z
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=66caca1e210445e38955d8cc5cdb3319@intel.com \
--to=qi.z.zhang@intel.com \
--cc=alvinx.zhang@intel.com \
--cc=dev@dpdk.org \
--cc=jia.guo@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).