DPDK patches and discussions
 help / color / mirror / Atom feed
From: Chaoyong He <chaoyong.he@corigine.com>
To: Ferruh Yigit <ferruh.yigit@amd.com>, "dev@dpdk.org" <dev@dpdk.org>
Cc: oss-drivers <oss-drivers@corigine.com>,
	Niklas Soderlund <niklas.soderlund@corigine.com>
Subject: RE: [PATCH v2 02/24] net/nfp: add the structures and functions for flow offload
Date: Wed, 19 Oct 2022 02:50:07 +0000	[thread overview]
Message-ID: <SJ0PR13MB554503D9387E62807F67CF949E2B9@SJ0PR13MB5545.namprd13.prod.outlook.com> (raw)
In-Reply-To: <39e84172-6c6a-06a1-523c-607a7cb4c25e@amd.com>

> On 10/10/2022 7:08 AM, Chaoyong He wrote:
> > Add the structures and functions to process mask table, flow table,
> > and flow stats id, which are used in the rte_flow offload logics.
> >
> > Signed-off-by: Chaoyong He <chaoyong.he@corigine.com>
> > Reviewed-by: Niklas Söderlund <niklas.soderlund@corigine.com>
> 
> <...>
> 
> > +__rte_unused static int
> > +nfp_flow_table_delete(struct nfp_flow_priv *priv,
> > +             struct rte_flow *nfp_flow) {
> > +     int ret;
> > +     char *hash_data;
> > +     uint32_t hash_key;
> > +
> > +     hash_data = (char *)(nfp_flow->payload.unmasked_data);
> > +     hash_key = rte_jhash(hash_data, nfp_flow->length,
> > + priv->hash_seed);
> 
> It can be an option to store hash information in 'struct rte_flow', instead of
> keep calculating it.
> 

Okay, I will revise like this in the next version patch, thanks.

> <...>
> 
> > +     /* flow stats */
> > +     rte_spinlock_init(&priv->stats_lock);
> > +     stats_size = (ctx_count & NFP_FL_STAT_ID_STAT) |
> > +                     ((ctx_split - 1) & NFP_FL_STAT_ID_MU_NUM);
> > +     PMD_INIT_LOG(INFO, "ctx_count:%0lx, ctx_split:%0lx, stats_size:%0lx
> ",
> > +                     ctx_count, ctx_split, stats_size);
> > +     priv->stats = rte_zmalloc("nfp_flow_stats",
> > +                     stats_size * sizeof(struct nfp_fl_stats), 0);
> 
> Is this allocates 'stats' for max possible flow rule count? Isn't this a too big
> number to allocate in advance?

For now, the 'stats_size' for our card is 500,000, and sizeof(structure nfp_fl_stats) is 16,
so the size of memory need alloc here is about 8MB. Maybe it's not a very big memory?

And we also calculate an index use these two macros(in function nfp_stats_id_alloc()),
and then send it to firmware, and the firmware will send the stats with this index to PMD.
If we alloc a small memory at start and then enlarge it at some point, we can't make sure 
the index we calculated is unique anymore, and the stats will confusion, which is not good.


  reply	other threads:[~2022-10-19  2:50 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-10  6:08 [PATCH v2 00/24] add the basic rte_flow offload support of nfp PMD Chaoyong He
2022-10-10  6:08 ` [PATCH v2 01/24] net/nfp: add the stats process logic in ctrl VNIC service Chaoyong He
2022-10-10 14:48   ` Ferruh Yigit
2022-10-10  6:08 ` [PATCH v2 02/24] net/nfp: add the structures and functions for flow offload Chaoyong He
2022-10-10 14:49   ` Ferruh Yigit
2022-10-19  2:50     ` Chaoyong He [this message]
2022-10-19 10:48       ` Ferruh Yigit
2022-10-10  6:08 ` [PATCH v2 03/24] net/nfp: add the flow APIs of nfp PMD Chaoyong He
2022-10-10 14:51   ` Ferruh Yigit
2022-10-19  3:00     ` Chaoyong He
2022-10-19 11:11       ` Ferruh Yigit
2022-10-19 11:30         ` Chaoyong He
2022-10-19 11:38           ` Ferruh Yigit
2022-10-10  6:08 ` [PATCH v2 04/24] net/nfp: add the offload support of basic items Chaoyong He
2022-10-10  6:08 ` [PATCH v2 05/24] net/nfp: add the offload support of basic actions Chaoyong He
2022-10-10 14:52   ` Ferruh Yigit
2022-10-19 11:32     ` Chaoyong He
2022-10-10  6:08 ` [PATCH v2 06/24] net/nfp: add the offload support of VLAN item Chaoyong He
2022-10-10  6:08 ` [PATCH v2 07/24] net/nfp: add the offload support of IPv4 item Chaoyong He
2022-10-10  6:08 ` [PATCH v2 08/24] net/nfp: add the offload support of IPv6 item Chaoyong He
2022-10-10 14:53   ` Ferruh Yigit
2022-10-19 11:33     ` Chaoyong He
2022-10-10  6:08 ` [PATCH v2 09/24] net/nfp: add the offload support of TCP item Chaoyong He
2022-10-10  6:08 ` [PATCH v2 10/24] net/nfp: add the offload support of UDP item Chaoyong He
2022-10-10  6:08 ` [PATCH v2 11/24] net/nfp: add the offload support of SCTP item Chaoyong He
2022-10-10  6:08 ` [PATCH v2 12/24] net/nfp: add the offload support of set SRC MAC action Chaoyong He
2022-10-10  6:08 ` [PATCH v2 13/24] net/nfp: add the offload support of set DST " Chaoyong He
2022-10-10  6:08 ` [PATCH v2 14/24] net/nfp: add the offload support of pop VLAN action Chaoyong He
2022-10-10  6:08 ` [PATCH v2 15/24] net/nfp: add the offload support of push " Chaoyong He
2022-10-10  6:08 ` [PATCH v2 16/24] net/nfp: add the offload support of set SRC IPv4 action Chaoyong He
2022-10-10  6:08 ` [PATCH v2 17/24] net/nfp: add the offload support of set DST " Chaoyong He
2022-10-10  6:08 ` [PATCH v2 18/24] net/nfp: add the offload support of set SRC IPv6 action Chaoyong He
2022-10-10  6:08 ` [PATCH v2 19/24] net/nfp: add the offload support of set DST " Chaoyong He
2022-10-10  6:08 ` [PATCH v2 20/24] net/nfp: add the offload support of set TP SRC port action Chaoyong He
2022-10-10  6:08 ` [PATCH v2 21/24] net/nfp: add the offload support of set TP DST " Chaoyong He
2022-10-10  6:09 ` [PATCH v2 22/24] net/nfp: add the offload support of set TTL action Chaoyong He
2022-10-10  6:09 ` [PATCH v2 23/24] net/nfp: add the offload support of set IPv4 DSCP action Chaoyong He
2022-10-10  6:09 ` [PATCH v2 24/24] net/nfp: add the offload support of set IPv6 " Chaoyong He

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=SJ0PR13MB554503D9387E62807F67CF949E2B9@SJ0PR13MB5545.namprd13.prod.outlook.com \
    --to=chaoyong.he@corigine.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@amd.com \
    --cc=niklas.soderlund@corigine.com \
    --cc=oss-drivers@corigine.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).