From: Jie Wang <jie.wang@intel.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Chas Williams <3chas3@gmail.com>
Cc: linux-usb@vger.kernel.org, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org,
linux-atm-general@lists.sourceforge.net,
accessrunner-general@lists.sourceforge.net,
syzbot+9b195c4f412ea5c4e56a@syzkaller.appspotmail.com,
Jie Wang <jie.wang@intel.com>
Subject: [PATCH v3] usb: atm: cxacru: fix NULL deref of atm_dev on sysfs writes
Date: Tue, 22 Sep 2026 09:29:53 +0000 [thread overview]
Message-ID: <20260922092953.4048507-1-jie.wang@intel.com> (raw)
In-Reply-To: <2026091012-nastiness-coveted-74e3@gregkh>
The adsl_state and adsl_config sysfs attributes are created via
dev_groups, so the driver core exposes them as soon as ->probe()
returns. At that point instance->atm_dev can still be NULL:
usbatm_heavy_init() waits only for the heavy-init kthread to start, and
atm_dev is assigned later, in usbatm_atm_init(). A write in that window
passes the "instance == NULL" check and then dereferences the NULL
atm_dev: atm_err()/atm_info() expand to
instance->usbatm->atm_dev->number, and a poll/start request also calls
cxacru_poll_status() directly, which touches atm_dev via
atm_dev_signal_change() and atm_dev->link_rate.
Log via usb_err()/usb_info() instead; they use the USB interface
device, which stays valid for the whole write. Reject a poll/start
request with -ENODEV before poll_state is advanced to CXPOLL_POLLING,
so that cxacru_atm_start() still starts polling once atm_dev is ready.
Fixes: e605c30977bb ("USB: atm: cxacru: convert to use dev_groups")
Assisted-by: Gemini:gemini-3.6-flash Gemini:gemini-3.1-pro-preview syzbot
Reported-by: syzbot+9b195c4f412ea5c4e56a@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=9b195c4f412ea5c4e56a
Tested-by: syzbot+9b195c4f412ea5c4e56a@syzkaller.appspotmail.com
Signed-off-by: Jie Wang <jie.wang@intel.com>
---
v3:
- Guard in adsl_state_store() instead of cxacru_poll_status(); the v2
guard left poll_state wedged at CXPOLL_POLLING with no work queued, so
polling never restarted (Sashiko).
- Restore the Assisted-by tag.
v2:
- Drop the ->atm_dev entry gate (an unsynchronized TOCTOU); remove the
->atm_dev dereference from the logging paths instead.
v1: https://lore.kernel.org/all/f3ef0a67-79d7-47da-be6e-7e82d92e68fe@mail.kernel.org/
v2: https://lore.kernel.org/all/20260908150203.2119844-1-jie.wang@intel.com/
drivers/usb/atm/cxacru.c | 17 +++++++++++++----
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/drivers/usb/atm/cxacru.c b/drivers/usb/atm/cxacru.c
index 636f7886fc260..501a51b27e466 100644
--- a/drivers/usb/atm/cxacru.c
+++ b/drivers/usb/atm/cxacru.c
@@ -356,7 +356,7 @@ static ssize_t adsl_state_store(struct device *dev,
if (!strcmp(str_cmd, "stop") || !strcmp(str_cmd, "restart")) {
ret = cxacru_cm(instance, CM_REQUEST_CHIP_ADSL_LINE_STOP, NULL, 0, NULL, 0);
if (ret < 0) {
- atm_err(instance->usbatm, "change adsl state:"
+ usb_err(instance->usbatm, "change adsl state:"
" CHIP_ADSL_LINE_STOP returned %d\n", ret);
ret = -EIO;
@@ -376,7 +376,7 @@ static ssize_t adsl_state_store(struct device *dev,
if (!strcmp(str_cmd, "start") || !strcmp(str_cmd, "restart")) {
ret = cxacru_cm(instance, CM_REQUEST_CHIP_ADSL_LINE_START, NULL, 0, NULL, 0);
if (ret < 0) {
- atm_err(instance->usbatm, "change adsl state:"
+ usb_err(instance->usbatm, "change adsl state:"
" CHIP_ADSL_LINE_START returned %d\n", ret);
ret = -EIO;
@@ -396,6 +396,15 @@ static ssize_t adsl_state_store(struct device *dev,
poll = -1;
}
+ /* cxacru_poll_status() below dereferences atm_dev, which may not be
+ * set up yet; reject before poll_state is advanced so that
+ * cxacru_atm_start() can still start polling once it is.
+ */
+ if (poll == CXPOLL_POLLING && !instance->usbatm->atm_dev) {
+ ret = -ENODEV;
+ poll = -1;
+ }
+
if (poll == CXPOLL_POLLING) {
mutex_lock(&instance->poll_state_serialize);
switch (instance->poll_state) {
@@ -481,7 +490,7 @@ static ssize_t adsl_config_store(struct device *dev,
ret = cxacru_cm(instance, CM_REQUEST_CARD_DATA_SET,
(u8 *) data, 4 + num * 8, NULL, 0);
if (ret < 0) {
- atm_err(instance->usbatm,
+ usb_err(instance->usbatm,
"set card data returned %d\n", ret);
return -EIO;
}
@@ -490,7 +499,7 @@ static ssize_t adsl_config_store(struct device *dev,
snprintf(log + tmp*12, 13, " %02x=%08x",
le32_to_cpu(data[tmp * 2 + 1]),
le32_to_cpu(data[tmp * 2 + 2]));
- atm_info(instance->usbatm, "config%s\n", log);
+ usb_info(instance->usbatm, "config%s\n", log);
num = 0;
}
}
--
2.43.0
next prev parent reply other threads:[~2026-09-22 1:44 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-31 14:40 [PATCH] usb: atm: cxacru: fix NULL pointer dereference on uninitialized atm_dev syzbot
2026-09-02 8:42 ` netdev-bot+sashiko
2026-09-08 15:02 ` [PATCH v2] usb: atm: cxacru: fix NULL deref of atm_dev on sysfs writes Jie Wang
2026-09-10 12:47 ` Greg Kroah-Hartman
2026-09-22 9:29 ` Jie Wang [this message]
2026-09-26 2:35 ` [PATCH v3] " Jakub Kicinski
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=20260922092953.4048507-1-jie.wang@intel.com \
--to=jie.wang@intel.com \
--cc=3chas3@gmail.com \
--cc=accessrunner-general@lists.sourceforge.net \
--cc=gregkh@linuxfoundation.org \
--cc=linux-atm-general@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=syzbot+9b195c4f412ea5c4e56a@syzkaller.appspotmail.com \
/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.