public inbox for dev@dpdk.org
 help / color / mirror / Atom feed
From: Kai Ji <kai.ji@intel.com>
To: dev@dpdk.org
Cc: gakhil@marvell.com, chiluka@marvell.com,
	Kai Ji <kai.ji@intel.com>,
	stable@dpdk.org
Subject: [dpdk-dev v1] crypto/qat: fix EC session segfault on missing pkey/pub key
Date: Wed, 25 Mar 2026 15:50:40 +0000	[thread overview]
Message-ID: <20260325155040.2922854-1-kai.ji@intel.com> (raw)

session_set_ec() checked qat_session->xform.ec.pkey.length for
the NULL-allocation guards, but that field is zero (from the
preceding memset) at that point, so the guards never fired.
When test_ecpm() created a session the xform had no pkey or
public-key fields set (ECPM only needs curve_id), so the
unchecked memcpy calls received a NULL destination and a
garbage-length source, causing a segfault.

Fix session_set_ec() to check xform->ec.*.length, the caller-
supplied value, before allocating and copying each optional
field (pkey, q.x, q.y), skipping the allocation entirely when
the length is zero.

Fix test_ecpm() to zero-initialise the xform with memset so
that uninitialised pkey/q fields are never passed to the driver.

Also add a QAT_LOG(ERR) in pick_curve() default case so that
unsupported curve IDs are surfaced immediately rather than
producing the misleading 'Unsupported xform type' message
from qat_asym_session_configure().

Fixes: 064ef1b098d1 (\"test/crypto: remove PMD-specific asym test suites\")

Cc: stable@dpdk.org

Signed-off-by: Kai Ji <kai.ji@intel.com>
---
 app/test/test_cryptodev_asym.c |  2 +-
 drivers/crypto/qat/qat_asym.c  | 58 ++++++++++++++++++----------------
 drivers/crypto/qat/qat_ec.h    |  1 +
 3 files changed, 32 insertions(+), 29 deletions(-)

diff --git a/app/test/test_cryptodev_asym.c b/app/test/test_cryptodev_asym.c
index 07e5eb5842..bf1a1fc417 100644
--- a/app/test/test_cryptodev_asym.c
+++ b/app/test/test_cryptodev_asym.c
@@ -1816,7 +1816,7 @@ test_ecpm(enum curve curve_id)
 	asym_op = op->asym;
 
 	/* Setup asym xform */
-	xform.next = NULL;
+	memset(&xform, 0, sizeof(xform));
 	xform.xform_type = RTE_CRYPTO_ASYM_XFORM_ECPM;
 	xform.ec.curve_id = input_params.curve;
 
diff --git a/drivers/crypto/qat/qat_asym.c b/drivers/crypto/qat/qat_asym.c
index 7a296cad6c..9276b86c0e 100644
--- a/drivers/crypto/qat/qat_asym.c
+++ b/drivers/crypto/qat/qat_asym.c
@@ -1492,36 +1492,38 @@ session_set_ec(struct qat_asym_session *qat_session,
 	uint8_t *q_x = xform->ec.q.x.data;
 	uint8_t *q_y = xform->ec.q.y.data;
 
-	qat_session->xform.ec.pkey.data =
-		rte_malloc(NULL, xform->ec.pkey.length, 0);
-	if (qat_session->xform.ec.pkey.length &&
-		qat_session->xform.ec.pkey.data == NULL)
-		return -ENOMEM;
-	qat_session->xform.ec.q.x.data = rte_malloc(NULL,
-		xform->ec.q.x.length, 0);
-	if (qat_session->xform.ec.q.x.length &&
-		qat_session->xform.ec.q.x.data == NULL) {
-		rte_free(qat_session->xform.ec.pkey.data);
-		return -ENOMEM;
+	if (xform->ec.pkey.length) {
+		qat_session->xform.ec.pkey.data =
+			rte_malloc(NULL, xform->ec.pkey.length, 0);
+		if (qat_session->xform.ec.pkey.data == NULL)
+			return -ENOMEM;
+		memcpy(qat_session->xform.ec.pkey.data, pkey,
+			xform->ec.pkey.length);
+		qat_session->xform.ec.pkey.length = xform->ec.pkey.length;
 	}
-	qat_session->xform.ec.q.y.data = rte_malloc(NULL,
-		xform->ec.q.y.length, 0);
-	if (qat_session->xform.ec.q.y.length &&
-		qat_session->xform.ec.q.y.data == NULL) {
-		rte_free(qat_session->xform.ec.pkey.data);
-		rte_free(qat_session->xform.ec.q.x.data);
-		return -ENOMEM;
+	if (xform->ec.q.x.length) {
+		qat_session->xform.ec.q.x.data = rte_malloc(NULL,
+			xform->ec.q.x.length, 0);
+		if (qat_session->xform.ec.q.x.data == NULL) {
+			rte_free(qat_session->xform.ec.pkey.data);
+			return -ENOMEM;
+		}
+		memcpy(qat_session->xform.ec.q.x.data, q_x,
+			xform->ec.q.x.length);
+		qat_session->xform.ec.q.x.length = xform->ec.q.x.length;
+	}
+	if (xform->ec.q.y.length) {
+		qat_session->xform.ec.q.y.data = rte_malloc(NULL,
+			xform->ec.q.y.length, 0);
+		if (qat_session->xform.ec.q.y.data == NULL) {
+			rte_free(qat_session->xform.ec.pkey.data);
+			rte_free(qat_session->xform.ec.q.x.data);
+			return -ENOMEM;
+		}
+		memcpy(qat_session->xform.ec.q.y.data, q_y,
+			xform->ec.q.y.length);
+		qat_session->xform.ec.q.y.length = xform->ec.q.y.length;
 	}
-
-	memcpy(qat_session->xform.ec.pkey.data, pkey,
-		xform->ec.pkey.length);
-	qat_session->xform.ec.pkey.length = xform->ec.pkey.length;
-	memcpy(qat_session->xform.ec.q.x.data, q_x,
-		xform->ec.q.x.length);
-	qat_session->xform.ec.q.x.length = xform->ec.q.x.length;
-	memcpy(qat_session->xform.ec.q.y.data, q_y,
-		xform->ec.q.y.length);
-	qat_session->xform.ec.q.y.length = xform->ec.q.y.length;
 	qat_session->xform.ec.curve_id = xform->ec.curve_id;
 
 	return 0;
diff --git a/drivers/crypto/qat/qat_ec.h b/drivers/crypto/qat/qat_ec.h
index 0e02722c18..653854abc4 100644
--- a/drivers/crypto/qat/qat_ec.h
+++ b/drivers/crypto/qat/qat_ec.h
@@ -274,6 +274,7 @@ pick_curve(const struct rte_crypto_asym_xform *xform)
 	case RTE_CRYPTO_EC_GROUP_SECP521R1:
 		return SECP521R1;
 	default:
+		QAT_LOG(ERR, "Unsupported curve id %d", xform->ec.curve_id);
 		return -1;
 	}
 }
-- 
2.43.0


             reply	other threads:[~2026-03-25 15:50 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-25 15:50 Kai Ji [this message]
2026-03-31  1:48 ` [dpdk-dev v1] crypto/qat: fix EC session segfault on missing pkey/pub key Jiang, YuX

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=20260325155040.2922854-1-kai.ji@intel.com \
    --to=kai.ji@intel.com \
    --cc=chiluka@marvell.com \
    --cc=dev@dpdk.org \
    --cc=gakhil@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox