From mboxrd@z Thu Jan  1 00:00:00 1970
From: Stephen Hemminger <stephen@networkplumber.org>
Subject: Re: rte_lpm with larger nexthops or another method?
Date: Mon, 22 Jun 2015 23:51:02 -0400
Message-ID: <20150622235102.41c3619a@uryu.home.lan>
References: <5A3882CB-0DE0-43DB-8DCA-051D561AA943@mhcomputing.net>
 <CANDrEHnuEwX6WGBFbMiB0C38m6axMRiAe3z=9OcjsZkGbPZSSA@mail.gmail.com>
 <20150622175302.GA15788@mhcomputing.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Cc: "<dev@dpdk.org>" <dev@dpdk.org>
To: Matthew Hall <mhall@mhcomputing.net>
Return-path: <dev-bounces@dpdk.org>
Received: from mail-pa0-f52.google.com (mail-pa0-f52.google.com
 [209.85.220.52]) by dpdk.org (Postfix) with ESMTP id AFCF1C74E
 for <dev@dpdk.org>; Tue, 23 Jun 2015 05:51:07 +0200 (CEST)
Received: by paceq1 with SMTP id eq1so122917007pac.3
 for <dev@dpdk.org>; Mon, 22 Jun 2015 20:51:07 -0700 (PDT)
In-Reply-To: <20150622175302.GA15788@mhcomputing.net>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

In order to make Vyatta/Brocade router work with LPM code
I ended up redoing the layout. It is:


/** Tbl24 entry structure. */
struct rte_lpm_tbl24_entry {
	/* Using single uint8_t to store 3 values. */
	uint8_t valid       :1; /**< Validation flag. */
	uint8_t ext_entry   :1; /**< external entry? */
	uint8_t depth;	      /**< Rule depth. */
	/* Stores Next hop or group index (i.e. gindex)into tbl8. */
	union {
		uint16_t next_hop;
		uint16_t tbl8_gindex;
	};
};

/** Tbl8 entry structure. */
struct rte_lpm_tbl8_entry {
	uint16_t next_hop;	/**< next hop. */
	uint8_t  depth;	/**< Rule depth. */
	uint8_t valid       :1; /**< Validation flag. */
	uint8_t valid_group :1; /**< Group validation flag. */
};

And also several other scalability improvements (plus IPv6)
and the correct handling of /32.

Unfortunately, this is such a big binary change that I was
reluctant to break any tests or applications using existing code
and therefore never submitted the patches.