All of lore.kernel.org
 help / color / mirror / Atom feed
From: Adam Casella <ihateprod@gmail.com>
To: netfilter@vger.kernel.org
Cc: Adam Casella <adam.casella1984@gmail.com>
Subject: [PATCH] conntrackd: cache: fix zone entry uniqueness in external cache
Date: Tue, 17 Aug 2021 13:31:25 -0700	[thread overview]
Message-ID: <20210817203125.20128-1-adam.casella1984@gmail.com> (raw)

In some use-cases, zone is used to differetiate conntrack state.  This preserves that uniqueness by adding zone into the cache in addtion to 5-tuple data
This preserves external-cache uniqueness per zone when synced.

Follow up fix to: https://git.netfilter.org/conntrack-tools/commit/?id=a08af5d26297eb85218a3c3a9e0991001a88cf10

Signed-off-by: Adam Casella <adam.casella1984@gmail.com>
---
 src/cache-ct.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/src/cache-ct.c b/src/cache-ct.c
index abcfde4..7e788d2 100644
--- a/src/cache-ct.c
+++ b/src/cache-ct.c
@@ -41,7 +41,8 @@ cache_hash4_ct(const struct nf_conntrack *ct, const struct hashtable *table)
 			  nfct_get_attr_u8(ct, ATTR_L4PROTO),
 		[3]	= nfct_get_attr_u16(ct, ATTR_PORT_SRC) << 16 |
 			  nfct_get_attr_u16(ct, ATTR_PORT_DST),
-	};
+                [4]     = nfct_get_attr_u16(ct, ATTR_ZONE),
+        };
 
 	/*
 	 * Instead of returning hash % table->hashsize (implying a divide)
@@ -50,13 +51,13 @@ cache_hash4_ct(const struct nf_conntrack *ct, const struct hashtable *table)
 	 * but using a multiply, less expensive than a divide. See:
 	 * http://www.mail-archive.com/netdev@vger.kernel.org/msg56623.html
 	 */
-	return ((uint64_t)jhash2(a, 4, 0) * table->hashsize) >> 32;
+	return ((uint64_t)jhash2(a, 5, 0) * table->hashsize) >> 32;
 }
 
 static uint32_t
 cache_hash6_ct(const struct nf_conntrack *ct, const struct hashtable *table)
 {
-	uint32_t a[10];
+	uint32_t a[11];
 
 	memcpy(&a[0], nfct_get_attr(ct, ATTR_IPV6_SRC), sizeof(uint32_t)*4);
 	memcpy(&a[4], nfct_get_attr(ct, ATTR_IPV6_DST), sizeof(uint32_t)*4);
@@ -64,8 +65,9 @@ cache_hash6_ct(const struct nf_conntrack *ct, const struct hashtable *table)
 	       nfct_get_attr_u8(ct, ATTR_ORIG_L4PROTO);
 	a[9] = nfct_get_attr_u16(ct, ATTR_ORIG_PORT_SRC) << 16 |
 	       nfct_get_attr_u16(ct, ATTR_ORIG_PORT_DST);
+        a[10] = nfct_get_attr_u16(ct, ATTR_ZONE);
 
-	return ((uint64_t)jhash2(a, 10, 0) * table->hashsize) >> 32;
+	return ((uint64_t)jhash2(a, 11, 0) * table->hashsize) >> 32;
 }
 
 static uint32_t
-- 
2.30.1 (Apple Git-130)


             reply	other threads:[~2021-08-17 20:31 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-17 20:31 Adam Casella [this message]
2021-09-20  7:56 ` [PATCH] conntrackd: cache: fix zone entry uniqueness in external cache Pablo Neira Ayuso

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=20210817203125.20128-1-adam.casella1984@gmail.com \
    --to=ihateprod@gmail.com \
    --cc=adam.casella1984@gmail.com \
    --cc=netfilter@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.