All of lore.kernel.org
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson@intel.com>
To: dev@dpdk.org
Cc: Bruce Richardson <bruce.richardson@intel.com>,
	stable@dpdk.org, Akhil Goyal <gakhil@marvell.com>,
	Anoob Joseph <anoobj@marvell.com>,
	Nithin Dabilpuram <ndabilpuram@marvell.com>
Subject: [PATCH 2/3] test/security_inline_proto: fix MTU calculation underflow
Date: Mon, 22 Jun 2026 12:18:33 +0100	[thread overview]
Message-ID: <20260622111835.233554-3-bruce.richardson@intel.com> (raw)
In-Reply-To: <20260622111835.233554-1-bruce.richardson@intel.com>

When testing with some NICs it was observed that the max MTU calculation
could underflow. Despite the RTE_MIN, due to integer promotion, this
lead to the configuring of the port with an excessively large, invalid
value. For example, if lim_nb_seg_max == 0, that meant that
max_data_room - 256 is -256 for comparison purposes using standard
integers (promoted from uint16_t). That is lower than the max mtu so the
-256 value is chosen, which becomes >65000 when converted back to
uint16_t.

Fix this by a) checking for seg_max == 0 b) doing calculations using
uint32_t and c) checking explicitly for underflow before subtracting.

Fixes: 3edd1197a605 ("test/security: add multi-segment inline IPsec cases")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 app/test/test_security_inline_proto.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/app/test/test_security_inline_proto.c b/app/test/test_security_inline_proto.c
index bc3ef54f71..81fce7364c 100644
--- a/app/test/test_security_inline_proto.c
+++ b/app/test/test_security_inline_proto.c
@@ -2045,8 +2045,19 @@ inline_ipsec_testsuite_setup(void)
 	memcpy(&local_port_conf, &port_conf, sizeof(port_conf));
 	/* Add Multi seg flags */
 	if (sg_mode) {
-		uint16_t max_data_room = RTE_MBUF_DEFAULT_DATAROOM *
-			dev_info.rx_desc_lim.nb_seg_max;
+		uint32_t max_data_room;
+
+		if (dev_info.rx_desc_lim.nb_seg_max == 0) {
+			printf("SG mode unsupported: invalid max Rx segments (0)\n");
+			return TEST_SKIPPED;
+		}
+
+		max_data_room = RTE_MBUF_DEFAULT_DATAROOM * dev_info.rx_desc_lim.nb_seg_max;
+		if (max_data_room <= 256) {
+			printf("SG mode unsupported: max data room (%u) too small\n",
+				max_data_room);
+			return TEST_SKIPPED;
+		}
 
 		local_port_conf.rxmode.offloads |= RTE_ETH_RX_OFFLOAD_SCATTER;
 		local_port_conf.txmode.offloads |= RTE_ETH_TX_OFFLOAD_MULTI_SEGS;
-- 
2.53.0


  parent reply	other threads:[~2026-06-22 11:18 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-22 11:18 [PATCH 0/3] Fixes for inline ipsec test cases Bruce Richardson
2026-06-22 11:18 ` [PATCH 1/3] test/security_inline_proto: remove fast-free Tx flag Bruce Richardson
2026-06-22 11:18 ` Bruce Richardson [this message]
2026-06-22 11:18 ` [PATCH 3/3] test/security_inline_proto: check for capabilities Bruce Richardson
2026-06-24 19:09 ` [EXTERNAL] [PATCH 0/3] Fixes for inline ipsec test cases Akhil Goyal

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=20260622111835.233554-3-bruce.richardson@intel.com \
    --to=bruce.richardson@intel.com \
    --cc=anoobj@marvell.com \
    --cc=dev@dpdk.org \
    --cc=gakhil@marvell.com \
    --cc=ndabilpuram@marvell.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 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.