This series fixes a refcount/locking imbalance in NFC LLCP receive handlers
when the socket is already in LLCP_CLOSED.
nfc_llcp_recv_disc() used to perform release_sock()/nfc_llcp_sock_put() in the CLOSED
branch but did not exit, and then performed the same cleanup again on the common
exit path. Drop the redundant CLOSED-branch cleanup so the common exit path runs
it exactly once, while keeping the existing DM_DISC reply behavior.
nfc_llcp_recv_hdlc() performed the CLOSED cleanup but then continued processing
and later cleaned up again on the common exit path. Return immediately after the
CLOSED cleanup.
Changes in v2:
- Drop Reported-by tags
- Add missing Fixes tags
Build-tested with: make M=net/nfc (no NFC HW available for runtime testing).
Qianchang Zhao (2):
nfc: llcp: avoid double release/put on LLCP_CLOSED in
nfc_llcp_recv_disc()
nfc: llcp: stop processing on LLCP_CLOSED in nfc_llcp_recv_hdlc()
net/nfc/llcp_core.c | 6 +-----
1 file changed, 1 insertion(+), 5 deletions(-)
--
2.34.1
If the IMX media pipeline is configured to receive multiple video
inputs, the second input stream may be broken on start. This happens if
the IMX CSI hardware has to be reconfigured for the second stream, while
the first stream is already running.
The IMX CSI driver configures the IMX CSI in the link_validate callback.
The media pipeline is only validated on the first start. Thus, any later
start of the media pipeline skips the validation and directly starts
streaming. This may leave the hardware in an inconsistent state compared
to the driver configuration. Moving the hardware configuration to the
stream start to make sure that the hardware is configured correctly.
Patch 1 removes the caching of the upstream mbus_config in
csi_link_validate and explicitly request the mbus_config in csi_start,
to get rid of this implicit dependency.
Patch 2 actually moves the hardware register setting from
csi_link_validate to csi_start to fix the skipped hardware
reconfiguration.
Signed-off-by: Michael Tretter <michael.tretter(a)pengutronix.de>
---
Changes in v2:
- Document changed locking in commit message
- Link to v1: https://lore.kernel.org/r/20251105-media-imx-fixes-v1-0-99e48b4f5cbc@pengut…
---
Michael Tretter (2):
media: staging: imx: request mbus_config in csi_start
media: staging: imx: configure src_mux in csi_start
drivers/staging/media/imx/imx-media-csi.c | 84 ++++++++++++++++++-------------
1 file changed, 48 insertions(+), 36 deletions(-)
---
base-commit: 27afd6e066cfd80ddbe22a4a11b99174ac89cced
change-id: 20251105-media-imx-fixes-acef77c7ba12
Best regards,
--
Michael Tretter <m.tretter(a)pengutronix.de>
From: Li Nan <linan122(a)huawei.com>
commit 9c47127a807da3e36ce80f7c83a1134a291fc021 upstream.
Raid checks if pad3 is zero when loading superblock from disk. Arrays
created with new features may fail to assemble on old kernels as pad3
is used.
Add module parameter check_new_feature to bypass this check.
Link: https://lore.kernel.org/linux-raid/20251103125757.1405796-5-linan666@huawei…
Signed-off-by: Li Nan <linan122(a)huawei.com>
Reviewed-by: Xiao Ni <xni(a)redhat.com>
Signed-off-by: Yu Kuai <yukuai(a)fnnas.com>
---
drivers/md/md.c | 12 +++++++++---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 4e033c26fdd4..9d9cb7e1e6e8 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -340,6 +340,7 @@ static int start_readonly;
*/
static bool create_on_open = true;
static bool legacy_async_del_gendisk = true;
+static bool check_new_feature = true;
/*
* We have a system wide 'event count' that is incremented
@@ -1752,9 +1753,13 @@ static int super_1_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_
}
if (sb->pad0 ||
sb->pad3[0] ||
- memcmp(sb->pad3, sb->pad3+1, sizeof(sb->pad3) - sizeof(sb->pad3[1])))
- /* Some padding is non-zero, might be a new feature */
- return -EINVAL;
+ memcmp(sb->pad3, sb->pad3+1, sizeof(sb->pad3) - sizeof(sb->pad3[1]))) {
+ pr_warn("Some padding is non-zero on %pg, might be a new feature\n",
+ rdev->bdev);
+ if (check_new_feature)
+ return -EINVAL;
+ pr_warn("check_new_feature is disabled, data corruption possible\n");
+ }
rdev->preferred_minor = 0xffff;
rdev->data_offset = le64_to_cpu(sb->data_offset);
@@ -10459,6 +10464,7 @@ module_param(start_dirty_degraded, int, S_IRUGO|S_IWUSR);
module_param_call(new_array, add_named_array, NULL, NULL, S_IWUSR);
module_param(create_on_open, bool, S_IRUSR|S_IWUSR);
module_param(legacy_async_del_gendisk, bool, 0600);
+module_param(check_new_feature, bool, 0600);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("MD RAID framework");
--
2.39.2
Originally, the notification for connector change will be enabled after
the first read of the connector status. Therefore, if the event happens
during this window, it will be missing and make the status unsynced.
Get the connector status only after enabling the notification for
connector change to ensure the status is synced.
Fixes: c1b0bc2dabfa ("usb: typec: Add support for UCSI interface")
Cc: stable(a)vger.kernel.org # v4.13+
Tested-by: Kenneth R. Crudup <kenny(a)panix.com>
Reviewed-by: Heikki Krogerus <heikki.krogerus(a)linux.intel.com>
Signed-off-by: Hsin-Te Yuan <yuanhsinte(a)chromium.org>
---
Changes in v7:
- Rebase onto 6.19-rc1
- Link to v6: https://lore.kernel.org/r/20251205-ucsi-v6-1-e2ad16550242@chromium.org
Changes in v6:
- Free the locks in error path.
- Link to v5: https://lore.kernel.org/r/20251205-ucsi-v5-1-488eb89bc9b8@chromium.org
Changes in v5:
- Hold the lock of each connector during the initialization to avoid
race condition between initialization and other event handler
- Add Fixes tag
- Link to v4: https://lore.kernel.org/r/20251125-ucsi-v4-1-8c94568ddaa5@chromium.org
Changes in v4:
- Handle a single connector in ucsi_init_port() and call it in a loop
- Link to v3: https://lore.kernel.org/r/20251121-ucsi-v3-1-b1047ca371b8@chromium.org
Changes in v3:
- Seperate the status checking part into a new function called
ucsi_init_port() and call it after enabling the notifications
- Link to v2: https://lore.kernel.org/r/20251118-ucsi-v2-1-d314d50333e2@chromium.org
Changes in v2:
- Remove unnecessary braces.
- Link to v1: https://lore.kernel.org/r/20251117-ucsi-v1-1-1dcbc5ea642b@chromium.org
---
drivers/usb/typec/ucsi/ucsi.c | 133 +++++++++++++++++++++++-------------------
1 file changed, 74 insertions(+), 59 deletions(-)
diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
index 9b3df776137a17e569588d41ee437a778342146e..f933c873f9e0d5f85275ebd96d34f90803230565 100644
--- a/drivers/usb/typec/ucsi/ucsi.c
+++ b/drivers/usb/typec/ucsi/ucsi.c
@@ -1624,11 +1624,71 @@ static struct fwnode_handle *ucsi_find_fwnode(struct ucsi_connector *con)
return NULL;
}
+static void ucsi_init_port(struct ucsi *ucsi, struct ucsi_connector *con)
+{
+ enum usb_role u_role = USB_ROLE_NONE;
+ int ret;
+
+ /* Get the status */
+ ret = ucsi_get_connector_status(con, false);
+ if (ret) {
+ dev_err(ucsi->dev, "con%d: failed to get status\n", con->num);
+ return;
+ }
+
+ if (ucsi->ops->connector_status)
+ ucsi->ops->connector_status(con);
+
+ switch (UCSI_CONSTAT(con, PARTNER_TYPE)) {
+ case UCSI_CONSTAT_PARTNER_TYPE_UFP:
+ case UCSI_CONSTAT_PARTNER_TYPE_CABLE_AND_UFP:
+ u_role = USB_ROLE_HOST;
+ fallthrough;
+ case UCSI_CONSTAT_PARTNER_TYPE_CABLE:
+ typec_set_data_role(con->port, TYPEC_HOST);
+ break;
+ case UCSI_CONSTAT_PARTNER_TYPE_DFP:
+ u_role = USB_ROLE_DEVICE;
+ typec_set_data_role(con->port, TYPEC_DEVICE);
+ break;
+ default:
+ break;
+ }
+
+ /* Check if there is already something connected */
+ if (UCSI_CONSTAT(con, CONNECTED)) {
+ typec_set_pwr_role(con->port, UCSI_CONSTAT(con, PWR_DIR));
+ ucsi_register_partner(con);
+ ucsi_pwr_opmode_change(con);
+ ucsi_orientation(con);
+ ucsi_port_psy_changed(con);
+ if (con->ucsi->cap.features & UCSI_CAP_GET_PD_MESSAGE)
+ ucsi_get_partner_identity(con);
+ if (con->ucsi->cap.features & UCSI_CAP_CABLE_DETAILS)
+ ucsi_check_cable(con);
+ }
+
+ /* Only notify USB controller if partner supports USB data */
+ if (!(UCSI_CONSTAT(con, PARTNER_FLAG_USB)))
+ u_role = USB_ROLE_NONE;
+
+ ret = usb_role_switch_set_role(con->usb_role_sw, u_role);
+ if (ret)
+ dev_err(ucsi->dev, "con:%d: failed to set usb role:%d\n",
+ con->num, u_role);
+
+ if (con->partner && UCSI_CONSTAT(con, PWR_OPMODE) == UCSI_CONSTAT_PWR_OPMODE_PD) {
+ ucsi_register_device_pdos(con);
+ ucsi_get_src_pdos(con);
+ ucsi_check_altmodes(con);
+ ucsi_check_connector_capability(con);
+ }
+}
+
static int ucsi_register_port(struct ucsi *ucsi, struct ucsi_connector *con)
{
struct typec_capability *cap = &con->typec_cap;
enum typec_accessory *accessory = cap->accessory;
- enum usb_role u_role = USB_ROLE_NONE;
u64 command;
char *name;
int ret;
@@ -1729,63 +1789,6 @@ static int ucsi_register_port(struct ucsi *ucsi, struct ucsi_connector *con)
goto out;
}
- /* Get the status */
- ret = ucsi_get_connector_status(con, false);
- if (ret) {
- dev_err(ucsi->dev, "con%d: failed to get status\n", con->num);
- goto out;
- }
-
- if (ucsi->ops->connector_status)
- ucsi->ops->connector_status(con);
-
- switch (UCSI_CONSTAT(con, PARTNER_TYPE)) {
- case UCSI_CONSTAT_PARTNER_TYPE_UFP:
- case UCSI_CONSTAT_PARTNER_TYPE_CABLE_AND_UFP:
- u_role = USB_ROLE_HOST;
- fallthrough;
- case UCSI_CONSTAT_PARTNER_TYPE_CABLE:
- typec_set_data_role(con->port, TYPEC_HOST);
- break;
- case UCSI_CONSTAT_PARTNER_TYPE_DFP:
- u_role = USB_ROLE_DEVICE;
- typec_set_data_role(con->port, TYPEC_DEVICE);
- break;
- default:
- break;
- }
-
- /* Check if there is already something connected */
- if (UCSI_CONSTAT(con, CONNECTED)) {
- typec_set_pwr_role(con->port, UCSI_CONSTAT(con, PWR_DIR));
- ucsi_register_partner(con);
- ucsi_pwr_opmode_change(con);
- ucsi_orientation(con);
- ucsi_port_psy_changed(con);
- if (con->ucsi->cap.features & UCSI_CAP_GET_PD_MESSAGE)
- ucsi_get_partner_identity(con);
- if (con->ucsi->cap.features & UCSI_CAP_CABLE_DETAILS)
- ucsi_check_cable(con);
- }
-
- /* Only notify USB controller if partner supports USB data */
- if (!(UCSI_CONSTAT(con, PARTNER_FLAG_USB)))
- u_role = USB_ROLE_NONE;
-
- ret = usb_role_switch_set_role(con->usb_role_sw, u_role);
- if (ret) {
- dev_err(ucsi->dev, "con:%d: failed to set usb role:%d\n",
- con->num, u_role);
- ret = 0;
- }
-
- if (con->partner && UCSI_CONSTAT(con, PWR_OPMODE) == UCSI_CONSTAT_PWR_OPMODE_PD) {
- ucsi_register_device_pdos(con);
- ucsi_get_src_pdos(con);
- ucsi_check_altmodes(con);
- ucsi_check_connector_capability(con);
- }
-
trace_ucsi_register_port(con->num, con);
out:
@@ -1903,17 +1906,29 @@ static int ucsi_init(struct ucsi *ucsi)
goto err_unregister;
}
+ /* Delay other interactions with each connector until ucsi_init_port is done */
+ for (i = 0; i < ucsi->cap.num_connectors; i++)
+ mutex_lock(&connector[i].lock);
+
/* Enable all supported notifications */
ntfy = ucsi_get_supported_notifications(ucsi);
command = UCSI_SET_NOTIFICATION_ENABLE | ntfy;
ucsi->message_in_size = 0;
ret = ucsi_send_command(ucsi, command);
- if (ret < 0)
+ if (ret < 0) {
+ for (i = 0; i < ucsi->cap.num_connectors; i++)
+ mutex_unlock(&connector[i].lock);
goto err_unregister;
+ }
ucsi->connector = connector;
ucsi->ntfy = ntfy;
+ for (i = 0; i < ucsi->cap.num_connectors; i++) {
+ ucsi_init_port(ucsi, &connector[i]);
+ mutex_unlock(&connector[i].lock);
+ }
+
mutex_lock(&ucsi->ppm_lock);
ret = ucsi->ops->read_cci(ucsi, &cci);
mutex_unlock(&ucsi->ppm_lock);
---
base-commit: ea1013c1539270e372fc99854bc6e4d94eaeff66
change-id: 20251117-ucsi-c2dfe8c006d7
Best regards,
--
Hsin-Te Yuan <yuanhsinte(a)chromium.org>