From: Wentao Liang <vulab@iscas.ac.cn>
To: daniel@zonque.org
Cc: dbrownell@users.sourceforge.net, gregkh@linuxfoundation.org,
haojian.zhuang@gmail.com, linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, linux-usb@vger.kernel.org,
robert.jarzmik@free.fr, Wentao Liang <vulab@iscas.ac.cn>,
stable@vger.kernel.org
Subject: [PATCH] usb: gadget: Fix transceiver leak in pxa_udc_probe()
Date: Wed, 16 Sep 2026 16:45:18 +0000 [thread overview]
Message-ID: <20260916164518.2084871-1-vulab@iscas.ac.cn> (raw)
When the device is probed from platform data instead of device tree,
pxa_udc_probe() takes a reference on the USB2 transceiver with
usb_get_phy(USB_PHY_TYPE_USB2). This reference is only released in
pxa_udc_remove(), but the error paths taken after the transceiver has
been acquired return directly: a failed D+ gpio lookup, a failed
devm_clk_get() or clk_prepare() call, a failed devm_request_irq() and
a failed usb_add_gadget_udc() or udc_enable() call all leak the
reference.
Route those error paths to the existing err label so that the
transceiver is put together with the clk_unprepare() cleanup. The put
is guarded by !of_have_populated_dt(), because in the device tree case
the reference is taken by devm_usb_get_phy_by_phandle() and released
by devres when the probe fails.
Fixes: 7fec3c25b773 ("USB: pxa27x_udc: add otg transceiver support")
Cc: stable@vger.kernel.org
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
---
drivers/usb/gadget/udc/pxa27x_udc.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/usb/gadget/udc/pxa27x_udc.c b/drivers/usb/gadget/udc/pxa27x_udc.c
index 1abea0d48c35..5f4f66e571f9 100644
--- a/drivers/usb/gadget/udc/pxa27x_udc.c
+++ b/drivers/usb/gadget/udc/pxa27x_udc.c
@@ -2421,18 +2421,21 @@ static int pxa_udc_probe(struct platform_device *pdev)
if (IS_ERR(udc->gpiod)) {
dev_err(&pdev->dev, "Couldn't find or request D+ gpio : %ld\n",
PTR_ERR(udc->gpiod));
- return PTR_ERR(udc->gpiod);
+ retval = PTR_ERR(udc->gpiod);
+ goto err;
}
if (udc->gpiod)
gpiod_direction_output(udc->gpiod, 0);
udc->clk = devm_clk_get(&pdev->dev, NULL);
- if (IS_ERR(udc->clk))
- return PTR_ERR(udc->clk);
+ if (IS_ERR(udc->clk)) {
+ retval = PTR_ERR(udc->clk);
+ goto err;
+ }
retval = clk_prepare(udc->clk);
if (retval)
- return retval;
+ goto err;
udc->vbus_sensed = 0;
@@ -2471,6 +2474,8 @@ static int pxa_udc_probe(struct platform_device *pdev)
usb_unregister_notifier(udc->transceiver, &pxa27x_udc_phy);
err:
clk_unprepare(udc->clk);
+ if (!of_have_populated_dt() && !IS_ERR_OR_NULL(udc->transceiver))
+ usb_put_phy(udc->transceiver);
return retval;
}
--
2.34.1
reply other threads:[~2026-09-16 16:46 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20260916164518.2084871-1-vulab@iscas.ac.cn \
--to=vulab@iscas.ac.cn \
--cc=daniel@zonque.org \
--cc=dbrownell@users.sourceforge.net \
--cc=gregkh@linuxfoundation.org \
--cc=haojian.zhuang@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=robert.jarzmik@free.fr \
--cc=stable@vger.kernel.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