DPDK patches and discussions
 help / color / mirror / Atom feed
From: Konstantin Ananyev <konstantin.ananyev@intel.com>
To: dev@dpdk.org
Cc: stable@dpdk.org, ryan.e.hall@intel.com,
	alexander.v.gutkin@intel.com,
	Konstantin Ananyev <konstantin.ananyev@intel.com>
Subject: [dpdk-dev] [PATCH 2/2] ip_frag: use key length for key comparision
Date: Mon,  5 Nov 2018 10:26:43 +0000	[thread overview]
Message-ID: <1541413603-4792-3-git-send-email-konstantin.ananyev@intel.com> (raw)
In-Reply-To: <1541413603-4792-1-git-send-email-konstantin.ananyev@intel.com>

Right now reassembly code relies on src_dst[] being all zeroes to
determine is it  free/occupied entry in the fragments table.
This is suboptimal and error prone - user can crash DPDK ip_reassembly
app by something like the following scapy script:
x=Ether(src=...,dst=...)/IP(dst='0.0.0.0',src='0.0.0.0',id=0)/('X'*1000)
frags=fragment(x, fragsize=500)
sendp(frags, iface=...)
To overcome that issue and reduce overhead of
'key invalidate'  and 'key is empty' operations -
add key_len into keys comparision procedure.

Reported-by: <ryan.e.hall@intel.com>
Reported-by: <alexander.v.gutkin@intel.com>

Fixes: 4f1a8f633862 ("ip_frag: add IPv6 reassembly")
Cc: stable@dpdk.org

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
 lib/librte_ip_frag/ip_frag_common.h | 14 ++++----------
 lib/librte_ip_frag/rte_ip_frag.h    | 14 +++++++++++---
 2 files changed, 15 insertions(+), 13 deletions(-)

diff --git a/lib/librte_ip_frag/ip_frag_common.h b/lib/librte_ip_frag/ip_frag_common.h
index 0f62e2e16..a17a74076 100644
--- a/lib/librte_ip_frag/ip_frag_common.h
+++ b/lib/librte_ip_frag/ip_frag_common.h
@@ -58,20 +58,14 @@ struct rte_mbuf *ipv6_frag_reassemble(struct ip_frag_pkt *fp);
 static inline int
 ip_frag_key_is_empty(const struct ip_frag_key * key)
 {
-	uint32_t i;
-	for (i = 0; i < RTE_MIN(key->key_len, RTE_DIM(key->src_dst)); i++)
-		if (key->src_dst[i] != 0)
-			return 0;
-	return 1;
+	return (key->key_len == 0);
 }
 
-/* empty the key */
+/* invalidate the key */
 static inline void
 ip_frag_key_invalidate(struct ip_frag_key * key)
 {
-	uint32_t i;
-	for (i = 0; i < key->key_len; i++)
-		key->src_dst[i] = 0;
+	key->key_len = 0;
 }
 
 /* compare two keys */
@@ -80,7 +74,7 @@ ip_frag_key_cmp(const struct ip_frag_key * k1, const struct ip_frag_key * k2)
 {
 	uint32_t i;
 	uint64_t val;
-	val = k1->id ^ k2->id;
+	val = k1->id_key_len ^ k2->id_key_len;
 	for (i = 0; i < k1->key_len; i++)
 		val |= k1->src_dst[i] ^ k2->src_dst[i];
 	return val;
diff --git a/lib/librte_ip_frag/rte_ip_frag.h b/lib/librte_ip_frag/rte_ip_frag.h
index 7f425f610..a4ccaf9d1 100644
--- a/lib/librte_ip_frag/rte_ip_frag.h
+++ b/lib/librte_ip_frag/rte_ip_frag.h
@@ -44,9 +44,17 @@ struct ip_frag {
 
 /** @internal <src addr, dst_addr, id> to uniquely identify fragmented datagram. */
 struct ip_frag_key {
-	uint64_t src_dst[4];      /**< src address, first 8 bytes used for IPv4 */
-	uint32_t id;           /**< dst address */
-	uint32_t key_len;      /**< src/dst key length */
+	uint64_t src_dst[4];
+	/**< src and dst address, only first 8 bytes used for IPv4 */
+	RTE_STD_C11
+	union {
+		uint64_t id_key_len; /**< combined for easy fetch */
+		__extension__
+		struct {
+			uint32_t id;       /**< packet id */
+			uint32_t key_len;  /**< src/dst key length */
+		};
+	};
 };
 
 /**
-- 
2.17.1

  parent reply	other threads:[~2018-11-05 10:27 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-05 10:26 [dpdk-dev] [PATCH 0/2] ip_frag: two fixes in reassembly code Konstantin Ananyev
2018-11-05 10:26 ` [dpdk-dev] [PATCH 1/2] ip_frag: check fragment length of incoming packet Konstantin Ananyev
2018-11-05 11:37   ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
2018-11-05 10:26 ` Konstantin Ananyev [this message]
2018-11-05 12:11 ` [dpdk-dev] [PATCH v2 0/2] ip_frag: two fixes in reassembly code Konstantin Ananyev
2018-11-05 12:11 ` [dpdk-dev] [PATCH v2 1/2] ip_frag: check fragment length of incoming packet Konstantin Ananyev
2018-11-05 12:11 ` [dpdk-dev] [PATCH 2/2] ip_frag: use key length for key comparision Konstantin Ananyev
2018-11-05 12:18 ` [dpdk-dev] [PATCH v2 0/2] ip_frag: two fixes in reassembly code Konstantin Ananyev
2018-11-06  0:58   ` Thomas Monjalon
2018-11-05 12:18 ` [dpdk-dev] [PATCH v2 1/2] ip_frag: check fragment length of incoming packet Konstantin Ananyev
2018-11-05 12:18 ` [dpdk-dev] [PATCH v2 2/2] ip_frag: use key length for key comparision Konstantin Ananyev
2018-11-06 10:53   ` Burakov, Anatoly
2018-11-06 11:41     ` Ananyev, Konstantin

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=1541413603-4792-3-git-send-email-konstantin.ananyev@intel.com \
    --to=konstantin.ananyev@intel.com \
    --cc=alexander.v.gutkin@intel.com \
    --cc=dev@dpdk.org \
    --cc=ryan.e.hall@intel.com \
    --cc=stable@dpdk.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 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).