* [PATCH v3 0/1] Fix and improve zcrypt reply message checks
@ 2026-09-10 16:19 Harald Freudenberger
2026-09-10 16:19 ` [PATCH v3 1/1] s390/zcrypt: Fix and improve zcrypt reply message verification checks Harald Freudenberger
0 siblings, 1 reply; 4+ messages in thread
From: Harald Freudenberger @ 2026-09-10 16:19 UTC (permalink / raw)
To: dengler, fcallies, ifranzki
Cc: freude, linux-s390, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev
Fix and improve zcrypt reply messages checks of type 6 - CCA and EP11
- and type 50 - ICA.
Changelog:
v1: Initial version
v2: Sashiko complained about not checking offset1/count1 at all if
count2 for an CCA type6 reply is given. And a check like this was
in place before the rework. So now offset1/count1 is always
checked and only if count2 is non zero offset2/count2 is
additionally checked.
v3: Again Sashiko had some complains which let to a broader review of
the reply processing functions. Especially the zcrypt_type50.c
functions convert_type80() and convert_response() got slight but
significant improvements - and also a mysterious BUG_ON has been
removed.
Harald Freudenberger (1):
s390/zcrypt: Fix and improve zcrypt reply message verification checks
drivers/s390/crypto/zcrypt_msgtype50.c | 95 ++++++++----
drivers/s390/crypto/zcrypt_msgtype6.c | 196 ++++++++++++++++++-------
2 files changed, 214 insertions(+), 77 deletions(-)
--
2.43.0
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH v3 1/1] s390/zcrypt: Fix and improve zcrypt reply message verification checks
2026-09-10 16:19 [PATCH v3 0/1] Fix and improve zcrypt reply message checks Harald Freudenberger
@ 2026-09-10 16:19 ` Harald Freudenberger
2026-09-10 16:35 ` sashiko-bot
0 siblings, 1 reply; 4+ messages in thread
From: Harald Freudenberger @ 2026-09-10 16:19 UTC (permalink / raw)
To: dengler, fcallies, ifranzki
Cc: freude, linux-s390, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev
Add or improve checks related to buffer sizes and reply sizes to the
handling of replies from the crypto cards for CCA, EP11 (AP message
type 6) and ICA (AP type 50) messages. The verification code related
to reply field length was not designed well and thus firmware
deficiencies could lead to unexpected behavior in the zcrypt device
driver. Thus improve the code to more closely inspect especially
length fields at message replies.
Add length validation before accessing reply message structures in
zcrypt_msgtype6_receive(), zcrypt_msgtype6_receive_ep11(), and
zcrypt_msgtype50_receive() to prevent out-of-bounds reads and
potential kernel memory disclosure.
Also rework these three functions to validate reply lengths more
carefully before copying data back into the request buffer. Use size_t
for length calculations, reject inconsistent reply sizes, and add
defensive handling for short invalid replies. For XCRB replies,
validate both reply segments and derive the effective message length
from the covered range instead of trusting only the second segment.
Additional improve some of the later invoked evaluation functions
which further check the payload. Add comments about length assumptions
and for the type50 processing rework the payload processing completely
with removing a misplaced BUG_ON().
Fixes: 3b6245fd303f ("s390/zcrypt: Separate msgtype implementation from card modules.")
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Cc: stable@vger.kernel.org
---
drivers/s390/crypto/zcrypt_msgtype50.c | 95 ++++++++----
drivers/s390/crypto/zcrypt_msgtype6.c | 196 ++++++++++++++++++-------
2 files changed, 214 insertions(+), 77 deletions(-)
diff --git a/drivers/s390/crypto/zcrypt_msgtype50.c b/drivers/s390/crypto/zcrypt_msgtype50.c
index d6fc2d8e7fad..5907fb2a6920 100644
--- a/drivers/s390/crypto/zcrypt_msgtype50.c
+++ b/drivers/s390/crypto/zcrypt_msgtype50.c
@@ -152,6 +152,11 @@ struct type80_hdr {
unsigned char reserved3[8];
} __packed;
+struct type80_reply {
+ struct type80_hdr hdr;
+ char data[];
+} __packed;
+
int get_rsa_modex_fc(struct ica_rsa_modexpo *mex, int *fcode)
{
if (!mex->inputdatalength)
@@ -340,32 +345,43 @@ static int ICACRT_msg_to_type50CRT_msg(struct zcrypt_queue *zq,
* @data: pointer to user output data
* @length: size of user output data
*
- * Returns 0 on success or -EFAULT.
+ * Returns 0 on success or neg. errno value on failure.
*/
static int convert_type80(struct zcrypt_queue *zq,
struct ap_message *reply,
char __user *outputdata,
unsigned int outputdatalength)
{
- struct type80_hdr *t80h = reply->msg;
- unsigned char *data;
+ struct type80_reply *msg = reply->msg;
+ size_t payload_len;
+
+ /*
+ * reply->len is always >= sizeof(struct type80_reply) here and
+ * outputdatalength (the modulus size) is guaranteed to be equal to
+ * inputdatalength and 0 < outputdatalength <= CEX3A_MAX_MOD_SIZE,
+ * also payload size may be >= outputdatalength.
+ */
+
+ payload_len = reply->len - sizeof(msg->hdr);
- if (t80h->len < sizeof(*t80h) + outputdatalength) {
- /* The result is too short, the CEXxA card may not do that.. */
+ if (outputdatalength > payload_len) {
+ /* We expect at least outputdatalength bytes, broken card ? */
zq->online = 0;
pr_err("Crypto dev=%02x.%04x code=0x%02x => online=0 rc=EAGAIN\n",
AP_QID_CARD(zq->queue->qid),
- AP_QID_QUEUE(zq->queue->qid), t80h->code);
+ AP_QID_QUEUE(zq->queue->qid), msg->hdr.code);
ZCRYPT_DBF_ERR("%s dev=%02x.%04x code=0x%02x => online=0 rc=EAGAIN\n",
__func__, AP_QID_CARD(zq->queue->qid),
- AP_QID_QUEUE(zq->queue->qid), t80h->code);
+ AP_QID_QUEUE(zq->queue->qid), msg->hdr.code);
ap_send_online_uevent(&zq->queue->ap_dev, zq->online);
return -EAGAIN;
}
- BUG_ON(t80h->len > CEX3A_MAX_RESPONSE_SIZE);
- data = reply->msg + t80h->len - outputdatalength;
- if (copy_to_user(outputdata, data, outputdatalength))
+
+ if (copy_to_user(outputdata,
+ msg->data + payload_len - outputdatalength,
+ outputdatalength))
return -EFAULT;
+
return 0;
}
@@ -374,14 +390,19 @@ static int convert_response(struct zcrypt_queue *zq,
char __user *outputdata,
unsigned int outputdatalength)
{
- /* Response type byte is the second byte in the response. */
- unsigned char rtype = ((unsigned char *)reply->msg)[1];
+ struct type80_reply *msg = reply->msg;
+
+ /* reply->len is always >= sizeof(struct error_hdr) here */
- switch (rtype) {
+ switch (msg->hdr.type) {
case TYPE82_RSP_CODE:
case TYPE88_RSP_CODE:
return convert_error(zq, reply);
case TYPE80_RSP_CODE:
+ if (msg->hdr.code)
+ return convert_error(zq, reply);
+ if (reply->len < sizeof(struct type80_reply))
+ return -EINVAL;
return convert_type80(zq, reply,
outputdata, outputdatalength);
default: /* Unknown response type, this should NEVER EVER happen */
@@ -389,11 +410,11 @@ static int convert_response(struct zcrypt_queue *zq,
pr_err("Crypto dev=%02x.%04x unknown response type 0x%02x => online=0 rc=EAGAIN\n",
AP_QID_CARD(zq->queue->qid),
AP_QID_QUEUE(zq->queue->qid),
- (int)rtype);
+ (int)msg->hdr.type);
ZCRYPT_DBF_ERR(
"%s dev=%02x.%04x unknown response type 0x%02x => online=0 rc=EAGAIN\n",
__func__, AP_QID_CARD(zq->queue->qid),
- AP_QID_QUEUE(zq->queue->qid), (int)rtype);
+ AP_QID_QUEUE(zq->queue->qid), (int)msg->hdr.type);
ap_send_online_uevent(&zq->queue->ap_dev, zq->online);
return -EAGAIN;
}
@@ -416,26 +437,46 @@ static void zcrypt_msgtype50_receive(struct ap_queue *aq,
.reply_code = REP82_ERROR_MACHINE_FAILURE,
};
struct type80_hdr *t80h;
- int len;
+ size_t len;
/* Copy the reply message to the request message buffer. */
if (!reply)
goto out; /* ap_msg->rc indicates the error */
+
t80h = reply->msg;
- if (t80h->type == TYPE80_RSP_CODE) {
- len = t80h->len;
- if (len > reply->bufsize || len > msg->bufsize ||
- len != reply->len) {
- pr_debug("len mismatch => EMSGSIZE\n");
- msg->rc = -EMSGSIZE;
+
+ if (reply->len < sizeof(*t80h) ||
+ t80h->type != TYPE80_RSP_CODE) {
+ if (reply->len < sizeof(error_reply)) {
+ /* total broken reply, use static error reply instead */
+ memcpy(msg->msg, &error_reply, sizeof(error_reply));
+ msg->len = sizeof(error_reply);
goto out;
+ } else {
+ /* malformed reply, convert function will handle this */
+ len = reply->len;
+ goto copy_len_and_out;
}
- memcpy(msg->msg, reply->msg, len);
- msg->len = len;
- } else {
- memcpy(msg->msg, reply->msg, sizeof(error_reply));
- msg->len = sizeof(error_reply);
}
+
+ len = t80h->len;
+
+copy_len_and_out:
+ if (len != reply->len) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("len %zu rpl.len %zu mismatch, msg.rc=%d\n",
+ len, reply->len, msg->rc);
+ goto out;
+ }
+ if (len > reply->bufsize || len > msg->bufsize) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("len %zu exceeds buf %zu/%zu, msg.rc=%d\n",
+ len, reply->bufsize, msg->bufsize, msg->rc);
+ goto out;
+ }
+ memcpy(msg->msg, reply->msg, len);
+ msg->len = len;
+
out:
complete(&msg->response.work);
}
diff --git a/drivers/s390/crypto/zcrypt_msgtype6.c b/drivers/s390/crypto/zcrypt_msgtype6.c
index 3df1d676de5d..a7d392272cfa 100644
--- a/drivers/s390/crypto/zcrypt_msgtype6.c
+++ b/drivers/s390/crypto/zcrypt_msgtype6.c
@@ -766,6 +766,15 @@ static int convert_type86_rng(struct zcrypt_queue *zq,
if (msg->cprbx.ccp_rtcode != 0 || msg->cprbx.ccp_rscode != 0)
return -EINVAL;
+ /*
+ * Note that offset2 and count2 have already been checked in
+ * zcrypt_msgtype6_receive(). So only check for valid count2
+ * and for not exceeding the hard coded rng buffer size.
+ */
+ if (!msg->fmt2.count2)
+ return -EINVAL;
+ if (msg->fmt2.count2 > ZCRYPT_RNG_BUFFER_SIZE)
+ return -EMSGSIZE;
memcpy(buffer, data + msg->fmt2.offset2, msg->fmt2.count2);
return msg->fmt2.count2;
}
@@ -777,11 +786,15 @@ static int convert_response_ica(struct zcrypt_queue *zq,
{
struct type86x_reply *msg = reply->msg;
+ /* reply->len is always >= sizeof(struct error_hdr) here */
+
switch (msg->hdr.type) {
case TYPE82_RSP_CODE:
case TYPE88_RSP_CODE:
return convert_error(zq, reply);
case TYPE86_RSP_CODE:
+ if (reply->len < sizeof(struct type86x_reply))
+ return -EINVAL;
if (msg->cprbx.ccp_rtcode &&
msg->cprbx.ccp_rscode == 0x14f &&
outputdatalength > 256) {
@@ -820,6 +833,8 @@ static int convert_response_xcrb(bool userspace, struct zcrypt_queue *zq,
{
struct type86x_reply *msg = reply->msg;
+ /* reply->len is always >= sizeof(struct error_hdr) here */
+
switch (msg->hdr.type) {
case TYPE82_RSP_CODE:
case TYPE88_RSP_CODE:
@@ -827,9 +842,14 @@ static int convert_response_xcrb(bool userspace, struct zcrypt_queue *zq,
return convert_error(zq, reply);
case TYPE86_RSP_CODE:
if (msg->hdr.reply_code) {
- xcrb->status = msg->fmt2.apfs;
+ if (reply->len < sizeof(struct type86_fmt2_msg))
+ xcrb->status = 0x0008044DL;
+ else
+ xcrb->status = msg->fmt2.apfs;
return convert_error(zq, reply);
}
+ if (reply->len < sizeof(struct type86x_reply))
+ return -EINVAL;
if (msg->cprbx.cprb_ver_id == 0x02)
return convert_type86_xcrb(userspace, zq, reply, xcrb);
fallthrough; /* wrong cprb version is an unknown response */
@@ -854,11 +874,15 @@ static int convert_response_ep11_xcrb(bool userspace, struct zcrypt_queue *zq,
{
struct type86_ep11_reply *msg = reply->msg;
+ /* reply->len is always >= sizeof(struct error_hdr) here */
+
switch (msg->hdr.type) {
case TYPE82_RSP_CODE:
case TYPE87_RSP_CODE:
return convert_error(zq, reply);
case TYPE86_RSP_CODE:
+ if (reply->len < sizeof(struct type86_ep11_reply))
+ return -EINVAL;
if (msg->hdr.reply_code)
return convert_error(zq, reply);
if (msg->cprbx.cprb_ver_id == 0x04)
@@ -885,6 +909,8 @@ static int convert_response_rng(struct zcrypt_queue *zq,
{
struct type86x_reply *msg = reply->msg;
+ /* reply->len is always >= sizeof(struct error_hdr) here */
+
switch (msg->hdr.type) {
case TYPE82_RSP_CODE:
case TYPE88_RSP_CODE:
@@ -892,6 +918,8 @@ static int convert_response_rng(struct zcrypt_queue *zq,
case TYPE86_RSP_CODE:
if (msg->hdr.reply_code)
return -EINVAL;
+ if (reply->len < sizeof(struct type86x_reply))
+ return -EINVAL;
if (msg->cprbx.cprb_ver_id == 0x02)
return convert_type86_rng(zq, reply, data);
fallthrough; /* wrong cprb version is an unknown response */
@@ -928,48 +956,85 @@ static void zcrypt_msgtype6_receive(struct ap_queue *aq,
};
struct ap_response_type *resp_type = &msg->response;
struct type86x_reply *t86r;
- int len;
+ size_t minlen, len;
/* Copy the reply message to the request message buffer. */
if (!reply)
goto out; /* ap_msg->rc indicates the error */
+
t86r = reply->msg;
- if (t86r->hdr.type == TYPE86_RSP_CODE &&
- t86r->cprbx.cprb_ver_id == 0x02) {
- switch (resp_type->type) {
- case CEXXC_RESPONSE_TYPE_ICA:
- len = sizeof(struct type86x_reply) + t86r->length;
- if (len > reply->bufsize || len > msg->bufsize ||
- len != reply->len) {
- pr_debug("len mismatch => EMSGSIZE\n");
- msg->rc = -EMSGSIZE;
- goto out;
- }
- memcpy(msg->msg, reply->msg, len);
- msg->len = len;
- break;
- case CEXXC_RESPONSE_TYPE_XCRB:
- if (t86r->fmt2.count2)
- len = t86r->fmt2.offset2 + t86r->fmt2.count2;
- else
- len = t86r->fmt2.offset1 + t86r->fmt2.count1;
- if (len > reply->bufsize || len > msg->bufsize ||
- len != reply->len) {
- pr_debug("len mismatch => EMSGSIZE\n");
+ minlen = sizeof(t86r->hdr) + sizeof(t86r->fmt2) +
+ offsetof(struct CPRBX, cprb_ver_id) +
+ sizeof(t86r->cprbx.cprb_ver_id);
+
+ if (reply->len < minlen ||
+ t86r->hdr.type != TYPE86_RSP_CODE ||
+ t86r->cprbx.cprb_ver_id != 0x02) {
+ if (reply->len < sizeof(error_reply)) {
+ /* total broken reply, use static error reply instead */
+ memcpy(msg->msg, &error_reply, sizeof(error_reply));
+ msg->len = sizeof(error_reply);
+ goto out;
+ } else {
+ /* malformed reply, convert function will handle this */
+ len = reply->len;
+ goto copy_len_and_out;
+ }
+ }
+
+ switch (resp_type->type) {
+ case CEXXC_RESPONSE_TYPE_ICA:
+ if (reply->len < sizeof(struct type86x_reply)) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("rpl.len %zu < struct type86_reply, msg.rc=%d\n",
+ reply->len, msg->rc);
+ goto out;
+ }
+ len = sizeof(struct type86x_reply) + (size_t)t86r->length;
+ break;
+ case CEXXC_RESPONSE_TYPE_XCRB:
+ len = (size_t)t86r->fmt2.offset1 + (size_t)t86r->fmt2.count1;
+ if (len > reply->len) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("offset1 %u count1 %u rpl.len %zu mismatch, msg.rc=%d\n",
+ t86r->fmt2.offset1, t86r->fmt2.count1,
+ reply->len, msg->rc);
+ goto out;
+ }
+ if (t86r->fmt2.count2) {
+ len = (size_t)t86r->fmt2.offset2 +
+ (size_t)t86r->fmt2.count2;
+ if (len > reply->len) {
msg->rc = -EMSGSIZE;
+ pr_debug("offset2 %u count2 %u rpl.len %zu mismatch, msg.rc=%d\n",
+ t86r->fmt2.offset2, t86r->fmt2.count2,
+ reply->len, msg->rc);
goto out;
}
- memcpy(msg->msg, reply->msg, len);
- msg->len = len;
- break;
- default:
- memcpy(msg->msg, &error_reply, sizeof(error_reply));
- msg->len = sizeof(error_reply);
}
- } else {
- memcpy(msg->msg, reply->msg, sizeof(error_reply));
+ break;
+ default:
+ memcpy(msg->msg, &error_reply, sizeof(error_reply));
msg->len = sizeof(error_reply);
+ goto out;
+ }
+
+copy_len_and_out:
+ if (len != reply->len) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("len %zu rpl.len %zu mismatch, msg.rc=%d\n",
+ len, reply->len, msg->rc);
+ goto out;
}
+ if (len > reply->bufsize || len > msg->bufsize) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("len %zu exceeds buf %zu/%zu, msg.rc=%d\n",
+ len, reply->bufsize, msg->bufsize, msg->rc);
+ goto out;
+ }
+ memcpy(msg->msg, reply->msg, len);
+ msg->len = len;
+
out:
complete(&resp_type->work);
}
@@ -992,34 +1057,65 @@ static void zcrypt_msgtype6_receive_ep11(struct ap_queue *aq,
};
struct ap_response_type *resp_type = &msg->response;
struct type86_ep11_reply *t86r;
- int len;
+ size_t minlen, len;
/* Copy the reply message to the request message buffer. */
if (!reply)
goto out; /* ap_msg->rc indicates the error */
+
t86r = reply->msg;
- if (t86r->hdr.type == TYPE86_RSP_CODE &&
- t86r->cprbx.cprb_ver_id == 0x04) {
- switch (resp_type->type) {
- case CEXXC_RESPONSE_TYPE_EP11:
- len = t86r->fmt2.offset1 + t86r->fmt2.count1;
- if (len > reply->bufsize || len > msg->bufsize ||
- len != reply->len) {
- pr_debug("len mismatch => EMSGSIZE\n");
- msg->rc = -EMSGSIZE;
- goto out;
- }
- memcpy(msg->msg, reply->msg, len);
- msg->len = len;
- break;
- default:
+ minlen = sizeof(t86r->hdr) + sizeof(t86r->fmt2) +
+ offsetof(struct ep11_cprb, cprb_ver_id) +
+ sizeof(t86r->cprbx.cprb_ver_id);
+
+ if (reply->len < minlen ||
+ t86r->hdr.type != TYPE86_RSP_CODE ||
+ t86r->cprbx.cprb_ver_id != 0x04) {
+ if (reply->len < sizeof(error_reply)) {
+ /* total broken reply, use static error reply instead */
memcpy(msg->msg, &error_reply, sizeof(error_reply));
msg->len = sizeof(error_reply);
+ goto out;
+ } else {
+ /* malformed reply, convert function will handle this */
+ len = reply->len;
+ goto copy_len_and_out;
}
- } else {
- memcpy(msg->msg, reply->msg, sizeof(error_reply));
+ }
+
+ switch (resp_type->type) {
+ case CEXXC_RESPONSE_TYPE_EP11:
+ len = (size_t)t86r->fmt2.offset1 + (size_t)t86r->fmt2.count1;
+ if (len > reply->len) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("offset1 %u count1 %u rpl.len %zu mismatch, msg.rc=%d\n",
+ t86r->fmt2.offset1, t86r->fmt2.count1,
+ reply->len, msg->rc);
+ goto out;
+ }
+ break;
+ default:
+ memcpy(msg->msg, &error_reply, sizeof(error_reply));
msg->len = sizeof(error_reply);
+ goto out;
+ }
+
+copy_len_and_out:
+ if (len != reply->len) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("len %zu rpl.len %zu mismatch, msg.rc=%d\n",
+ len, reply->len, msg->rc);
+ goto out;
}
+ if (len > reply->bufsize || len > msg->bufsize) {
+ msg->rc = -EMSGSIZE;
+ pr_debug("len %zu exceeds buf %zu/%zu, msg.rc=%d\n",
+ len, reply->bufsize, msg->bufsize, msg->rc);
+ goto out;
+ }
+ memcpy(msg->msg, reply->msg, len);
+ msg->len = len;
+
out:
complete(&resp_type->work);
}
--
2.43.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH v3 1/1] s390/zcrypt: Fix and improve zcrypt reply message verification checks
2026-09-10 16:19 ` [PATCH v3 1/1] s390/zcrypt: Fix and improve zcrypt reply message verification checks Harald Freudenberger
@ 2026-09-10 16:35 ` sashiko-bot
2026-09-11 12:52 ` Harald Freudenberger
0 siblings, 1 reply; 4+ messages in thread
From: sashiko-bot @ 2026-09-10 16:35 UTC (permalink / raw)
To: Harald Freudenberger
Cc: linux-s390, Heiko Carstens, Vasily Gorbik, Alexander Gordeev,
Christian Borntraeger
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Legitimate XCRB and RNG responses will be spuriously rejected due to enforcing an incorrect minimum length of sizeof(struct type86x_reply).
--
commit 991141a9c29feb41fbfdd37e1610a682977e72b6
Author: Harald Freudenberger <freude@linux.ibm.com>
s390/zcrypt: Fix and improve zcrypt reply message verification checks
This commit adds and improves checks related to buffer sizes and reply sizes
for CCA, EP11, and ICA messages in the zcrypt driver. It adds length
validation before accessing reply message structures to prevent out-of-bounds
reads. It also reworks payload evaluation functions.
> diff --git a/drivers/s390/crypto/zcrypt_msgtype6.c b/drivers/s390/crypto/zcrypt_msgtype6.c
> index 3df1d676de5d8..a7d392272cfa2 100644
> --- a/drivers/s390/crypto/zcrypt_msgtype6.c
> +++ b/drivers/s390/crypto/zcrypt_msgtype6.c
[ ... ]
> @@ -820,6 +833,8 @@ static int convert_response_xcrb(bool userspace, struct zcrypt_queue *zq,
> {
> struct type86x_reply *msg = reply->msg;
>
> + /* reply->len is always >= sizeof(struct error_hdr) here */
> +
> switch (msg->hdr.type) {
> case TYPE82_RSP_CODE:
> case TYPE88_RSP_CODE:
> @@ -827,9 +842,14 @@ static int convert_response_xcrb(bool userspace, struct zcrypt_queue *zq,
> return convert_error(zq, reply);
> case TYPE86_RSP_CODE:
> if (msg->hdr.reply_code) {
> - xcrb->status = msg->fmt2.apfs;
> + if (reply->len < sizeof(struct type86_fmt2_msg))
> + xcrb->status = 0x0008044DL;
> + else
> + xcrb->status = msg->fmt2.apfs;
> return convert_error(zq, reply);
> }
> + if (reply->len < sizeof(struct type86x_reply))
> + return -EINVAL;
[Severity: High]
Does this code incorrectly reject legitimate XCRB responses?
It appears that this check assumes XCRB replies use the large ICA-specific
type86x_reply struct. If a userspace application sends a valid XCRB request
without a data payload, the hardware replies successfully without padding.
While zcrypt_msgtype6_receive() correctly computes len as offset1 + count1
and verifies it exactly matches the hardware reply length, this check evaluates
reply->len < sizeof(struct type86x_reply). Since type86x_reply includes 6
additional bytes (pad[4] and length) intended only for ICA responses, its size
is larger than an empty-payload XCRB response.
Could this result in valid XCRB responses being rejected with -EINVAL,
causing a regression for userspace applications using the ICAXCRB ioctl?
> if (msg->cprbx.cprb_ver_id == 0x02)
> return convert_type86_xcrb(userspace, zq, reply, xcrb);
> fallthrough; /* wrong cprb version is an unknown response */
[ ... ]
> @@ -885,6 +909,8 @@ static int convert_response_rng(struct zcrypt_queue *zq,
> {
> struct type86x_reply *msg = reply->msg;
>
> + /* reply->len is always >= sizeof(struct error_hdr) here */
> +
> switch (msg->hdr.type) {
> case TYPE82_RSP_CODE:
> case TYPE88_RSP_CODE:
> @@ -892,6 +918,8 @@ static int convert_response_rng(struct zcrypt_queue *zq,
> case TYPE86_RSP_CODE:
> if (msg->hdr.reply_code)
> return -EINVAL;
> + if (reply->len < sizeof(struct type86x_reply))
> + return -EINVAL;
[Severity: High]
Does this check similarly apply a logically flawed minimum-size limit to RNG
hardware responses?
It seems this could fail if the hardware RNG response buffer size
configuration ever drops below the 6-byte delta caused by the ICA-specific
padding in type86x_reply.
> if (msg->cprbx.cprb_ver_id == 0x02)
> return convert_type86_rng(zq, reply, data);
> fallthrough; /* wrong cprb version is an unknown response */
--
Sashiko AI review · https://sashiko.dev/#/patchset/20260910161906.36251-2-freude@linux.ibm.com?part=1
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3 1/1] s390/zcrypt: Fix and improve zcrypt reply message verification checks
2026-09-10 16:35 ` sashiko-bot
@ 2026-09-11 12:52 ` Harald Freudenberger
0 siblings, 0 replies; 4+ messages in thread
From: Harald Freudenberger @ 2026-09-11 12:52 UTC (permalink / raw)
To: sashiko-reviews
Cc: linux-s390, Heiko Carstens, Vasily Gorbik, Alexander Gordeev,
Christian Borntraeger
On 2026-09-10 18:35, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential
> issue(s) to consider:
> - [High] Legitimate XCRB and RNG responses will be spuriously rejected
> due to enforcing an incorrect minimum length of sizeof(struct
> type86x_reply).
> --
>
> commit 991141a9c29feb41fbfdd37e1610a682977e72b6
> Author: Harald Freudenberger <freude@linux.ibm.com>
>
> s390/zcrypt: Fix and improve zcrypt reply message verification checks
>
> This commit adds and improves checks related to buffer sizes and reply
> sizes
> for CCA, EP11, and ICA messages in the zcrypt driver. It adds length
> validation before accessing reply message structures to prevent
> out-of-bounds
> reads. It also reworks payload evaluation functions.
>
>> diff --git a/drivers/s390/crypto/zcrypt_msgtype6.c
>> b/drivers/s390/crypto/zcrypt_msgtype6.c
>> index 3df1d676de5d8..a7d392272cfa2 100644
>> --- a/drivers/s390/crypto/zcrypt_msgtype6.c
>> +++ b/drivers/s390/crypto/zcrypt_msgtype6.c
> [ ... ]
>> @@ -820,6 +833,8 @@ static int convert_response_xcrb(bool userspace,
>> struct zcrypt_queue *zq,
>> {
>> struct type86x_reply *msg = reply->msg;
>>
>> + /* reply->len is always >= sizeof(struct error_hdr) here */
>> +
>> switch (msg->hdr.type) {
>> case TYPE82_RSP_CODE:
>> case TYPE88_RSP_CODE:
>> @@ -827,9 +842,14 @@ static int convert_response_xcrb(bool userspace,
>> struct zcrypt_queue *zq,
>> return convert_error(zq, reply);
>> case TYPE86_RSP_CODE:
>> if (msg->hdr.reply_code) {
>> - xcrb->status = msg->fmt2.apfs;
>> + if (reply->len < sizeof(struct type86_fmt2_msg))
>> + xcrb->status = 0x0008044DL;
>> + else
>> + xcrb->status = msg->fmt2.apfs;
>> return convert_error(zq, reply);
>> }
>> + if (reply->len < sizeof(struct type86x_reply))
>> + return -EINVAL;
>
> [Severity: High]
> Does this code incorrectly reject legitimate XCRB responses?
>
> It appears that this check assumes XCRB replies use the large
> ICA-specific
> type86x_reply struct. If a userspace application sends a valid XCRB
> request
> without a data payload, the hardware replies successfully without
> padding.
>
> While zcrypt_msgtype6_receive() correctly computes len as offset1 +
> count1
> and verifies it exactly matches the hardware reply length, this check
> evaluates
> reply->len < sizeof(struct type86x_reply). Since type86x_reply includes
> 6
> additional bytes (pad[4] and length) intended only for ICA responses,
> its size
> is larger than an empty-payload XCRB response.
>
> Could this result in valid XCRB responses being rejected with -EINVAL,
> causing a regression for userspace applications using the ICAXCRB
> ioctl?
>
Will fix this in the next version.
>> if (msg->cprbx.cprb_ver_id == 0x02)
>> return convert_type86_xcrb(userspace, zq, reply, xcrb);
>> fallthrough; /* wrong cprb version is an unknown response */
> [ ... ]
>> @@ -885,6 +909,8 @@ static int convert_response_rng(struct
>> zcrypt_queue *zq,
>> {
>> struct type86x_reply *msg = reply->msg;
>>
>> + /* reply->len is always >= sizeof(struct error_hdr) here */
>> +
>> switch (msg->hdr.type) {
>> case TYPE82_RSP_CODE:
>> case TYPE88_RSP_CODE:
>> @@ -892,6 +918,8 @@ static int convert_response_rng(struct
>> zcrypt_queue *zq,
>> case TYPE86_RSP_CODE:
>> if (msg->hdr.reply_code)
>> return -EINVAL;
>> + if (reply->len < sizeof(struct type86x_reply))
>> + return -EINVAL;
>
> [Severity: High]
> Does this check similarly apply a logically flawed minimum-size limit
> to RNG
> hardware responses?
>
> It seems this could fail if the hardware RNG response buffer size
> configuration ever drops below the 6-byte delta caused by the
> ICA-specific
> padding in type86x_reply.
>
Does not apply. A valid RNG reply always need to be >= sizeof(struct
type86x_reply)
and even more. However to access the fields cprbx.ccp_rtcode and
cprbx.ccp_rscode in
function convert_type86_rng() this check is a precondition.
>> if (msg->cprbx.cprb_ver_id == 0x02)
>> return convert_type86_rng(zq, reply, data);
>> fallthrough; /* wrong cprb version is an unknown response */
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-09-11 12:52 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-09-10 16:19 [PATCH v3 0/1] Fix and improve zcrypt reply message checks Harald Freudenberger
2026-09-10 16:19 ` [PATCH v3 1/1] s390/zcrypt: Fix and improve zcrypt reply message verification checks Harald Freudenberger
2026-09-10 16:35 ` sashiko-bot
2026-09-11 12:52 ` Harald Freudenberger
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox