DPDK patches and discussions
 help / color / mirror / Atom feed
From: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
To: dev@dpdk.org
Cc: Yipeng Wang <yipeng1.wang@intel.com>,
	Sameh Gobriel <sameh.gobriel@intel.com>,
	Bruce Richardson <bruce.richardson@intel.com>
Subject: [RFC 1/4] thash: add RSS hash key generation API
Date: Fri,  6 Sep 2024 16:53:15 +0000	[thread overview]
Message-ID: <20240906165318.1322550-2-vladimir.medvedkin@intel.com> (raw)
In-Reply-To: <20240906165318.1322550-1-vladimir.medvedkin@intel.com>

Currently the only way to have non static Toeplitz hash key is to
generate it randomly. Such a key may not guarantee good packets
distribution, especially if there are small number of flows.

This patch adds stub implementation of the Toeplitz hash key generation
function, which may improve Toeplitz hash key with respect to hash values
distribution.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
---
 lib/hash/rte_thash.c | 13 +++++++++++++
 lib/hash/rte_thash.h | 24 ++++++++++++++++++++++++
 lib/hash/version.map |  4 +++-
 3 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/lib/hash/rte_thash.c b/lib/hash/rte_thash.c
index 10721effe6..df9b4960ce 100644
--- a/lib/hash/rte_thash.c
+++ b/lib/hash/rte_thash.c
@@ -832,3 +832,16 @@ rte_thash_adjust_tuple(struct rte_thash_ctx *ctx,
 
 	return ret;
 }
+
+int
+rte_thash_gen_key(uint8_t *key, int key_len, int reta_sz_log,
+	int entropy_start, int entropy_sz)
+{
+	RTE_SET_USED(key);
+	RTE_SET_USED(key_len);
+	RTE_SET_USED(reta_sz_log);
+	RTE_SET_USED(entropy_start);
+	RTE_SET_USED(entropy_sz);
+
+	return 0;
+}
diff --git a/lib/hash/rte_thash.h b/lib/hash/rte_thash.h
index 30b657e67a..560fc1de73 100644
--- a/lib/hash/rte_thash.h
+++ b/lib/hash/rte_thash.h
@@ -447,6 +447,30 @@ rte_thash_adjust_tuple(struct rte_thash_ctx *ctx,
 	uint32_t desired_value, unsigned int attempts,
 	rte_thash_check_tuple_t fn, void *userdata);
 
+/**
+ * @warning
+ * @b EXPERIMENTAL: this API may change without prior notice
+ *
+ * Modify RSS hash key such that subtuple bits corresponding to `entropy_sz`
+ * bits starting from `entropy_start` will have the most even distribution with
+ * this key with a given ReTa size.
+ *
+ * @param key pointer to the RSS hash key
+ * @param key_len length of the key
+ * @param reta_sz_log log2 of the size of RSS redirection table. i.e. number of
+ *   bits of the rss hash value used to identify RSS ReTa entry
+ * @param entropy_start bit offset from the beginning of the tuple where user
+ *   expects best distribution of the subtuple values.
+ * @param entropy_sz size in bits of the part of subtuple
+ *
+ * @return
+ *  0 on success negative otherwise
+ */
+__rte_experimental
+int
+rte_thash_gen_key(uint8_t *key, int key_len, int reta_sz_log,
+	int entropy_start, int entropy_sz);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/lib/hash/version.map b/lib/hash/version.map
index 11a5394a45..7d31fc1ba5 100644
--- a/lib/hash/version.map
+++ b/lib/hash/version.map
@@ -52,6 +52,8 @@ EXPERIMENTAL {
 
 	# added in 24.07
 	rte_hash_rcu_qsbr_dq_reclaim;
+	# added in 24.11
+	rte_thash_gen_key;
 };
 
 INTERNAL {
@@ -59,4 +61,4 @@ INTERNAL {
 
 	rte_thash_gfni_stub;
 	rte_thash_gfni_bulk_stub;
-};
+};
\ No newline at end of file
-- 
2.34.1


  reply	other threads:[~2024-09-06 16:53 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-06 16:53 [RFC 0/4] RSS hash key generation Vladimir Medvedkin
2024-09-06 16:53 ` Vladimir Medvedkin [this message]
2024-09-09  0:09   ` [RFC 1/4] thash: add RSS hash key generation API Stephen Hemminger
2024-09-06 16:53 ` [RFC 2/4] hash: add dynamic polynomial calculation Vladimir Medvedkin
2024-09-09  0:11   ` Stephen Hemminger
2024-09-06 16:53 ` [RFC 3/4] hash: implement RSS hash key generation API Vladimir Medvedkin
2024-09-06 16:53 ` [RFC 4/4] test/thash: add tests for RSS " Vladimir Medvedkin

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=20240906165318.1322550-2-vladimir.medvedkin@intel.com \
    --to=vladimir.medvedkin@intel.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=sameh.gobriel@intel.com \
    --cc=yipeng1.wang@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).