From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Jeff Guo <jia.guo@intel.com>,
qi.z.zhang@intel.com, qiming.yang@intel.com
Cc: dev@dpdk.org, jingjing.wu@intel.com, junfeng.guo@intel.com,
beilei.xing@intel.com, simei.su@intel.com
Subject: Re: [dpdk-dev] [dpdk-dev v5 1/4] net/ice: refactor for pf hash flow
Date: Tue, 7 Jul 2020 15:24:57 +0100 [thread overview]
Message-ID: <d9d4e9f1-a0a5-c15f-dc81-46c9e8e7aef9@intel.com> (raw)
In-Reply-To: <20200707052546.14151-2-jia.guo@intel.com>
On 7/7/2020 6:25 AM, Jeff Guo wrote:
> Refactor hash flow by change the action parser and clean some code.
>
> Signed-off-by: Jeff Guo <jia.guo@intel.com>
<...>
> @@ -333,46 +507,24 @@ ice_hash_parse_action(struct ice_pattern_match_item *pattern_match_item,
> switch (action_type) {
> case RTE_FLOW_ACTION_TYPE_RSS:
> rss = action->conf;
> - rss_hf = rss->types;
> -
> - /**
> - * Check simultaneous use of SRC_ONLY and DST_ONLY
> - * of the same level.
> - */
> - rss_hf = rte_eth_rss_hf_refine(rss_hf);
> + rss_type = rss->types;
>
> - /* Check if pattern is empty. */
> + /* Check hash function and save it to rss_meta. */
> if (pattern_match_item->pattern_list !=
> - pattern_empty && rss->func ==
> - RTE_ETH_HASH_FUNCTION_SIMPLE_XOR)
> + pattern_empty && rss->func ==
> + RTE_ETH_HASH_FUNCTION_SIMPLE_XOR) {
> return rte_flow_error_set(error, ENOTSUP,
> RTE_FLOW_ERROR_TYPE_ACTION, action,
> "Not supported flow");
> -
> - if ((rss_hf & ETH_RSS_ETH) && (rss_hf & ~ETH_RSS_PPPOE))
> - m->eth_rss_hint = ETH_RSS_ETH;
> - else if ((rss_hf & ETH_RSS_PPPOE) && (rss_hf & ~ETH_RSS_ETH))
> - m->eth_rss_hint = ETH_RSS_PPPOE;
> - else if ((rss_hf & ETH_RSS_ETH) && (rss_hf & ETH_RSS_PPPOE))
> - m->eth_rss_hint = ETH_RSS_ETH | ETH_RSS_PPPOE;
> -
> - /* Check if rss types match pattern. */
> - if (rss->func != RTE_ETH_HASH_FUNCTION_SIMPLE_XOR) {
> - if (((rss_hf & ETH_RSS_IPV4) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_NONFRAG_IPV4_UDP) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_NONFRAG_IPV4_TCP) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_NONFRAG_IPV4_SCTP) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_IPV6) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_NONFRAG_IPV6_UDP) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_NONFRAG_IPV6_TCP) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_NONFRAG_IPV6_SCTP) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_ETH) != m->eth_rss_hint) &&
> - ((rss_hf & ETH_RSS_PPPOE) != m->eth_rss_hint) &&
> - (((rss_hf & (ETH_RSS_ETH | ETH_RSS_PPPOE)) !=
> - m->eth_rss_hint)))
> - return rte_flow_error_set(error,
> - ENOTSUP, RTE_FLOW_ERROR_TYPE_ACTION,
> - action, "Not supported RSS types");
> + } else if (rss->func ==
> + RTE_ETH_HASH_FUNCTION_SIMPLE_XOR){
> + ((struct rss_meta *)*meta)->hash_function =
> + RTE_ETH_HASH_FUNCTION_SIMPLE_XOR;
> + return 0;
> + } else if (rss->func ==
> + RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ) {
> + ((struct rss_meta *)*meta)->hash_function =
> + RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ;
> }
>
> if (rss->level)
> @@ -390,42 +542,38 @@ ice_hash_parse_action(struct ice_pattern_match_item *pattern_match_item,
> RTE_FLOW_ERROR_TYPE_ACTION, action,
> "a non-NULL RSS queue is not supported");
>
> - /* Check hash function and save it to rss_meta. */
> - if (rss->func ==
> - RTE_ETH_HASH_FUNCTION_SIMPLE_XOR)
> - ((struct rss_meta *)*meta)->hash_function =
> - RTE_ETH_HASH_FUNCTION_SIMPLE_XOR;
> + /**
> + * Check simultaneous use of SRC_ONLY and DST_ONLY
> + * of the same level.
> + */
> + rss_type = rte_eth_rss_hf_refine(rss_type);
>
> - if (rss->func ==
> - RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ)
> - ((struct rss_meta *)*meta)->hash_function =
> - RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ;
> + uint64_t combine_type = ETH_RSS_L2_SRC_ONLY |
> + ETH_RSS_L2_DST_ONLY |
> + ETH_RSS_L3_SRC_ONLY |
> + ETH_RSS_L3_DST_ONLY |
> + ETH_RSS_L4_SRC_ONLY |
> + ETH_RSS_L4_DST_ONLY;
Declaring and initializing a varible in switch case causing warning with icc
[1], I will fix this while mergin by moving declaration out of switch.
[1]
.../dpdk/drivers/net/ice/ice_hash.c(672):
warning #589: transfer of control bypasses initialization of:
variable "combine_type" (declared at line 716)
switch (action_type) {
^
<...>
next prev parent reply other threads:[~2020-07-07 14:25 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-12 2:57 [dpdk-dev] net/ice: enable new input set for rss hash Jeff Guo
2020-06-14 15:08 ` [dpdk-dev] [dpdk-dev v2] " Jeff Guo
2020-06-21 14:09 ` [dpdk-dev] [dpdk-dev v3 1/2] app/testpmd: add GTPU to RSS hash commands Jeff Guo
2020-06-21 14:09 ` [dpdk-dev] [dpdk-dev v3 2/2] net/ice: enable new input set for rss hash Jeff Guo
2020-07-01 13:02 ` Zhang, Qi Z
2020-07-03 3:53 ` Jeff Guo
2020-07-03 3:58 ` [dpdk-dev] [dpdk-dev v4 0/4] enable new hash flow for pf Jeff Guo
2020-07-03 3:58 ` [dpdk-dev] [dpdk-dev v4 1/4] net/ice: refactor for pf hash flow Jeff Guo
2020-07-03 3:58 ` [dpdk-dev] [dpdk-dev v4 2/4] net/ice: support hash for new GTPU protocols Jeff Guo
2020-07-03 3:58 ` [dpdk-dev] [dpdk-dev v4 3/4] net/ice: enable new input set for rss hash Jeff Guo
2020-07-03 3:58 ` [dpdk-dev] [dpdk-dev v4 4/4] app/testpmd: add GTPU to RSS hash commands Jeff Guo
2020-07-07 5:25 ` [dpdk-dev] [dpdk-dev v5 0/4] enable new hash flow for pf Jeff Guo
2020-07-07 5:25 ` [dpdk-dev] [dpdk-dev v5 1/4] net/ice: refactor for pf hash flow Jeff Guo
2020-07-07 14:24 ` Ferruh Yigit [this message]
2020-07-07 5:25 ` [dpdk-dev] [dpdk-dev v5 2/4] net/ice: support hash for new GTPU protocols Jeff Guo
2020-07-07 5:25 ` [dpdk-dev] [dpdk-dev v5 3/4] net/ice: enable new input set for rss hash Jeff Guo
2020-07-07 5:25 ` [dpdk-dev] [dpdk-dev v5 4/4] app/testpmd: add GTPU to RSS hash commands Jeff Guo
2020-07-07 9:31 ` [dpdk-dev] [dpdk-dev v5 0/4] enable new hash flow for pf 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=d9d4e9f1-a0a5-c15f-dc81-46c9e8e7aef9@intel.com \
--to=ferruh.yigit@intel.com \
--cc=beilei.xing@intel.com \
--cc=dev@dpdk.org \
--cc=jia.guo@intel.com \
--cc=jingjing.wu@intel.com \
--cc=junfeng.guo@intel.com \
--cc=qi.z.zhang@intel.com \
--cc=qiming.yang@intel.com \
--cc=simei.su@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).