From: "Cao, Yahui" <yahui.cao@intel.com>
To: "Yan, Zhirun" <zhirun.yan@intel.com>,
"Zhang, Qi Z" <qi.z.zhang@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>
Cc: "Wang, Xiao W" <xiao.w.wang@intel.com>,
"Su, Simei" <simei.su@intel.com>,
"Guo, Junfeng" <junfeng.guo@intel.com>
Subject: Re: [dpdk-dev] [PATCH v1 2/2] net/ice: support inner/outer L2/L3 field for FDIR
Date: Thu, 15 Oct 2020 09:28:40 +0000 [thread overview]
Message-ID: <9dbce9fbf43d47868c2534530108bac3@intel.com> (raw)
In-Reply-To: <20200928063146.668003-3-zhirun.yan@intel.com>
> -----Original Message-----
> From: Yan, Zhirun <zhirun.yan@intel.com>
> Sent: Monday, September 28, 2020 2:32 PM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>; dev@dpdk.org
> Cc: Cao, Yahui <yahui.cao@intel.com>; Wang, Xiao W
> <xiao.w.wang@intel.com>; Su, Simei <simei.su@intel.com>; Guo, Junfeng
> <junfeng.guo@intel.com>; Yan, Zhirun <zhirun.yan@intel.com>
> Subject: [PATCH v1 2/2] net/ice: support inner/outer L2/L3 field for FDIR
>
> Distinguish inner/outer fields for parse pattern. So FDIR for tunnel
> can be more flexible. Enable VXLAN inner/outer L3/L4 different fields
> for FDIR.
>
> Signed-off-by: Zhirun Yan <zhirun.yan@intel.com>
> ---
> drivers/net/ice/ice_fdir_filter.c | 59 +++++++++++++++++++++++++++++--
> 1 file changed, 56 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/ice/ice_fdir_filter.c b/drivers/net/ice/ice_fdir_filter.c
> index 175abcdd5..961528d17 100644
> --- a/drivers/net/ice/ice_fdir_filter.c
> +++ b/drivers/net/ice/ice_fdir_filter.c
> @@ -56,6 +56,11 @@
> ICE_INSET_SCTP_SRC_PORT | ICE_INSET_SCTP_DST_PORT)
>
> #define ICE_FDIR_INSET_VXLAN_IPV4 (\
> + ICE_FDIR_INSET_ETH | \
> + ICE_INSET_IPV4_SRC | ICE_INSET_IPV4_DST | \
> + ICE_INSET_IPV4_TOS | \
> + ICE_INSET_UDP_DST_PORT | \
> + ICE_INSET_TUN_DMAC | ICE_INSET_TUN_SMAC | \
> ICE_INSET_TUN_IPV4_SRC | ICE_INSET_TUN_IPV4_DST)
>
> #define ICE_FDIR_INSET_VXLAN_IPV4_TCP (\
> @@ -907,6 +912,7 @@ ice_fdir_input_set_parse(uint64_t inset, enum
> ice_flow_field *field)
> };
> static const struct ice_inset_map ice_inset_map[] = {
> {ICE_INSET_DMAC, ICE_FLOW_FIELD_IDX_ETH_DA},
> + {ICE_INSET_SMAC, ICE_FLOW_FIELD_IDX_ETH_SA},
> {ICE_INSET_ETHERTYPE, ICE_FLOW_FIELD_IDX_ETH_TYPE},
> {ICE_INSET_IPV4_SRC, ICE_FLOW_FIELD_IDX_IPV4_SA},
> {ICE_INSET_IPV4_DST, ICE_FLOW_FIELD_IDX_IPV4_DA},
> @@ -1655,6 +1661,14 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad,
> uint32_t vtc_flow_cpu;
> uint16_t ether_type;
> enum rte_flow_item_type next_type;
> + bool is_outer_part = true;
> +
> + for (item = pattern; item->type != RTE_FLOW_ITEM_TYPE_END; item++)
> {
> + if (item->type == RTE_FLOW_ITEM_TYPE_VXLAN) {
> + tunnel_type = ICE_FDIR_TUNNEL_TYPE_VXLAN;
> + break;
> + }
> + }
>
> for (item = pattern; item->type != RTE_FLOW_ITEM_TYPE_END; item++)
> {
> if (item->last) {
> @@ -1672,7 +1686,25 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad,
> eth_mask = item->mask;
> next_type = (item + 1)->type;
>
> - if (eth_spec && eth_mask) {
> + if (!(eth_spec && eth_mask))
> + break;
> +
> + /* handle outer L2 fields */
> + if (is_outer_part && tunnel_type == ICE_FDIR_TUNNEL_TYPE_VXLAN) {
> + if (!rte_is_zero_ether_addr(ð_mask->dst))
> {
> + filter->outer_input_set |= ICE_INSET_DMAC;
> + rte_memcpy(&filter->input.ext_data_outer.dst_mac,
> + ð_spec->dst,
> + RTE_ETHER_ADDR_LEN);
> + }
> +
> + if (!rte_is_zero_ether_addr(ð_mask->src)) {
> + filter->outer_input_set |=ICE_INSET_SMAC;
> + rte_memcpy(&filter->input.ext_data_outer.src_mac,
> + ð_spec->src,
> + RTE_ETHER_ADDR_LEN);
> + }
> + } else {
> if (!rte_is_zero_ether_addr(ð_mask->dst))
> {
> input_set |= ICE_INSET_DMAC;
> rte_memcpy(&filter->input.ext_data.dst_mac,
> @@ -1714,7 +1746,27 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad,
> ipv4_spec = item->spec;
> ipv4_mask = item->mask;
>
> - if (ipv4_spec && ipv4_mask) {
> + if (!(ipv4_spec && ipv4_mask))
> + break;
> +
[Cao, Yahui]
It seems below code has quite some duplication with code within the else condition
Can you use a pointer *input_set and *ip to point to the different input_set bits and input_set value,
So that existing code can be reused.
> + /* handle outer L3 fields */
> + if (is_outer_part && tunnel_type == ICE_FDIR_TUNNEL_TYPE_VXLAN) {
> + if (ipv4_mask->hdr.dst_addr == UINT32_MAX) {
> + filter->outer_input_set |= ICE_INSET_IPV4_DST;
> + filter->input.ip_outer.v4.dst_ip =
> + ipv4_spec->hdr.dst_addr;
> + }
> + if (ipv4_mask->hdr.src_addr == UINT32_MAX) {
> + filter->outer_input_set |= ICE_INSET_IPV4_SRC;
> + filter->input.ip_outer.v4.src_ip =
> + ipv4_spec->hdr.src_addr;
> + }
> + if (ipv4_mask->hdr.type_of_service == UINT8_MAX) {
> + input_set |= ICE_INSET_IPV4_TOS;
[Cao, Yahui]
Why use input_set instead of filter->outer_input_set in the VXLAN outer section ?
> + filter->input.ip_outer.v4.tos =
> + ipv4_spec->hdr.type_of_service;
> + }
> + } else {
> /* Check IPv4 mask and update input set */
> if (ipv4_mask->hdr.version_ihl ||
> ipv4_mask->hdr.total_length ||
> @@ -1944,6 +1996,8 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad,
> break;
> case RTE_FLOW_ITEM_TYPE_VXLAN:
> l3 = RTE_FLOW_ITEM_TYPE_END;
> + is_outer_part = false;
> +
> vxlan_spec = item->spec;
> vxlan_mask = item->mask;
>
> @@ -1955,7 +2009,6 @@ ice_fdir_parse_pattern(__rte_unused struct ice_adapter *ad,
> return -rte_errno;
> }
>
> - tunnel_type = ICE_FDIR_TUNNEL_TYPE_VXLAN;
> break;
> case RTE_FLOW_ITEM_TYPE_GTPU:
> l3 = RTE_FLOW_ITEM_TYPE_END;
> --
> 2.25.1
next prev parent reply other threads:[~2020-10-15 9:28 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-28 6:31 [dpdk-dev] [PATCH v1 0/2] support FDIR inner/outer field for tunnel packet Zhirun Yan
2020-09-28 6:31 ` [dpdk-dev] [PATCH v1 1/2] net/ice/base: add tunnel support for FDIR Zhirun Yan
2020-09-28 6:31 ` [dpdk-dev] [PATCH v1 2/2] net/ice: support inner/outer L2/L3 field " Zhirun Yan
2020-10-15 9:28 ` Cao, Yahui [this message]
2020-11-02 8:23 ` Yan, Zhirun
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=9dbce9fbf43d47868c2534530108bac3@intel.com \
--to=yahui.cao@intel.com \
--cc=dev@dpdk.org \
--cc=junfeng.guo@intel.com \
--cc=qi.z.zhang@intel.com \
--cc=simei.su@intel.com \
--cc=xiao.w.wang@intel.com \
--cc=zhirun.yan@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).