public inbox for stable@vger.kernel.org
 help / color / mirror / Atom feed
* FAILED: patch "[PATCH] USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels" failed to apply to 5.15-stable tree
@ 2025-09-17  7:44 gregkh
  2025-09-17 13:17 ` [PATCH 5.15.y 1/2] usb: gadget: dummy_hcd: remove usage of list iterator past the loop body Sasha Levin
  0 siblings, 1 reply; 3+ messages in thread
From: gregkh @ 2025-09-17  7:44 UTC (permalink / raw)
  To: stern, bigeasy, gregkh, stable, ysk; +Cc: stable


The patch below does not apply to the 5.15-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@vger.kernel.org>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y
git checkout FETCH_HEAD
git cherry-pick -x 8d63c83d8eb922f6c316320f50c82fa88d099bea
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '2025091723-stack-cargo-2b1d@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^..

Possible dependencies:



thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 8d63c83d8eb922f6c316320f50c82fa88d099bea Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 25 Aug 2025 12:00:22 -0400
Subject: [PATCH] USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels

Yunseong Kim and the syzbot fuzzer both reported a problem in
RT-enabled kernels caused by the way dummy-hcd mixes interrupt
management and spin-locking.  The pattern was:

	local_irq_save(flags);
	spin_lock(&dum->lock);
	...
	spin_unlock(&dum->lock);
	...		// calls usb_gadget_giveback_request()
	local_irq_restore(flags);

The code was written this way because usb_gadget_giveback_request()
needs to be called with interrupts disabled and the private lock not
held.

While this pattern works fine in non-RT kernels, it's not good when RT
is enabled.  RT kernels handle spinlocks much like mutexes; in particular,
spin_lock() may sleep.  But sleeping is not allowed while local
interrupts are disabled.

To fix the problem, rewrite the code to conform to the pattern used
elsewhere in dummy-hcd and other UDC drivers:

	spin_lock_irqsave(&dum->lock, flags);
	...
	spin_unlock(&dum->lock);
	usb_gadget_giveback_request(...);
	spin_lock(&dum->lock);
	...
	spin_unlock_irqrestore(&dum->lock, flags);

This approach satisfies the RT requirements.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@kernel.org>
Fixes: b4dbda1a22d2 ("USB: dummy-hcd: disable interrupts during req->complete")
Reported-by: Yunseong Kim <ysk@kzalloc.com>
Closes: <https://lore.kernel.org/linux-usb/5b337389-73b9-4ee4-a83e-7e82bf5af87a@kzalloc.com/>
Reported-by: syzbot+8baacc4139f12fa77909@syzkaller.appspotmail.com
Closes: <https://lore.kernel.org/linux-usb/68ac2411.050a0220.37038e.0087.GAE@google.com/>
Tested-by: syzbot+8baacc4139f12fa77909@syzkaller.appspotmail.com
CC: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
CC: stable@vger.kernel.org
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/bb192ae2-4eee-48ee-981f-3efdbbd0d8f0@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index 21dbfb0b3bac..1cefca660773 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -765,8 +765,7 @@ static int dummy_dequeue(struct usb_ep *_ep, struct usb_request *_req)
 	if (!dum->driver)
 		return -ESHUTDOWN;
 
-	local_irq_save(flags);
-	spin_lock(&dum->lock);
+	spin_lock_irqsave(&dum->lock, flags);
 	list_for_each_entry(iter, &ep->queue, queue) {
 		if (&iter->req != _req)
 			continue;
@@ -776,15 +775,16 @@ static int dummy_dequeue(struct usb_ep *_ep, struct usb_request *_req)
 		retval = 0;
 		break;
 	}
-	spin_unlock(&dum->lock);
 
 	if (retval == 0) {
 		dev_dbg(udc_dev(dum),
 				"dequeued req %p from %s, len %d buf %p\n",
 				req, _ep->name, _req->length, _req->buf);
+		spin_unlock(&dum->lock);
 		usb_gadget_giveback_request(_ep, _req);
+		spin_lock(&dum->lock);
 	}
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&dum->lock, flags);
 	return retval;
 }
 


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* [PATCH 5.15.y 1/2] usb: gadget: dummy_hcd: remove usage of list iterator past the loop body
  2025-09-17  7:44 FAILED: patch "[PATCH] USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels" failed to apply to 5.15-stable tree gregkh
@ 2025-09-17 13:17 ` Sasha Levin
  2025-09-17 13:17   ` [PATCH 5.15.y 2/2] USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels Sasha Levin
  0 siblings, 1 reply; 3+ messages in thread
From: Sasha Levin @ 2025-09-17 13:17 UTC (permalink / raw)
  To: stable; +Cc: Jakob Koschel, Greg Kroah-Hartman, Sasha Levin

From: Jakob Koschel <jakobkoschel@gmail.com>

[ Upstream commit 7975f080d3557725160a878b1a64339043ba3d91 ]

To move the list iterator variable into the list_for_each_entry_*()
macro in the future it should be avoided to use the list iterator
variable after the loop body.

To *never* use the list iterator variable after the loop it was
concluded to use a separate iterator variable [1].

Link: https://lore.kernel.org/all/YhdfEIwI4EdtHdym@kroah.com/
Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
Link: https://lore.kernel.org/r/20220308171818.384491-26-jakobkoschel@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Stable-dep-of: 8d63c83d8eb9 ("USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/usb/gadget/udc/dummy_hcd.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index a2d956af42a23..35aec8e7fc735 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -751,7 +751,7 @@ static int dummy_dequeue(struct usb_ep *_ep, struct usb_request *_req)
 	struct dummy		*dum;
 	int			retval = -EINVAL;
 	unsigned long		flags;
-	struct dummy_request	*req = NULL;
+	struct dummy_request	*req = NULL, *iter;
 
 	if (!_ep || !_req)
 		return retval;
@@ -763,13 +763,14 @@ static int dummy_dequeue(struct usb_ep *_ep, struct usb_request *_req)
 
 	local_irq_save(flags);
 	spin_lock(&dum->lock);
-	list_for_each_entry(req, &ep->queue, queue) {
-		if (&req->req == _req) {
-			list_del_init(&req->queue);
-			_req->status = -ECONNRESET;
-			retval = 0;
-			break;
-		}
+	list_for_each_entry(iter, &ep->queue, queue) {
+		if (&iter->req != _req)
+			continue;
+		list_del_init(&iter->queue);
+		_req->status = -ECONNRESET;
+		req = iter;
+		retval = 0;
+		break;
 	}
 	spin_unlock(&dum->lock);
 
-- 
2.51.0


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* [PATCH 5.15.y 2/2] USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels
  2025-09-17 13:17 ` [PATCH 5.15.y 1/2] usb: gadget: dummy_hcd: remove usage of list iterator past the loop body Sasha Levin
@ 2025-09-17 13:17   ` Sasha Levin
  0 siblings, 0 replies; 3+ messages in thread
From: Sasha Levin @ 2025-09-17 13:17 UTC (permalink / raw)
  To: stable
  Cc: Alan Stern, stable, Yunseong Kim, syzbot+8baacc4139f12fa77909,
	Sebastian Andrzej Siewior, Greg Kroah-Hartman, Sasha Levin

From: Alan Stern <stern@rowland.harvard.edu>

[ Upstream commit 8d63c83d8eb922f6c316320f50c82fa88d099bea ]

Yunseong Kim and the syzbot fuzzer both reported a problem in
RT-enabled kernels caused by the way dummy-hcd mixes interrupt
management and spin-locking.  The pattern was:

	local_irq_save(flags);
	spin_lock(&dum->lock);
	...
	spin_unlock(&dum->lock);
	...		// calls usb_gadget_giveback_request()
	local_irq_restore(flags);

The code was written this way because usb_gadget_giveback_request()
needs to be called with interrupts disabled and the private lock not
held.

While this pattern works fine in non-RT kernels, it's not good when RT
is enabled.  RT kernels handle spinlocks much like mutexes; in particular,
spin_lock() may sleep.  But sleeping is not allowed while local
interrupts are disabled.

To fix the problem, rewrite the code to conform to the pattern used
elsewhere in dummy-hcd and other UDC drivers:

	spin_lock_irqsave(&dum->lock, flags);
	...
	spin_unlock(&dum->lock);
	usb_gadget_giveback_request(...);
	spin_lock(&dum->lock);
	...
	spin_unlock_irqrestore(&dum->lock, flags);

This approach satisfies the RT requirements.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@kernel.org>
Fixes: b4dbda1a22d2 ("USB: dummy-hcd: disable interrupts during req->complete")
Reported-by: Yunseong Kim <ysk@kzalloc.com>
Closes: <https://lore.kernel.org/linux-usb/5b337389-73b9-4ee4-a83e-7e82bf5af87a@kzalloc.com/>
Reported-by: syzbot+8baacc4139f12fa77909@syzkaller.appspotmail.com
Closes: <https://lore.kernel.org/linux-usb/68ac2411.050a0220.37038e.0087.GAE@google.com/>
Tested-by: syzbot+8baacc4139f12fa77909@syzkaller.appspotmail.com
CC: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
CC: stable@vger.kernel.org
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/bb192ae2-4eee-48ee-981f-3efdbbd0d8f0@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/usb/gadget/udc/dummy_hcd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c
index 35aec8e7fc735..da658d349c9c3 100644
--- a/drivers/usb/gadget/udc/dummy_hcd.c
+++ b/drivers/usb/gadget/udc/dummy_hcd.c
@@ -761,8 +761,7 @@ static int dummy_dequeue(struct usb_ep *_ep, struct usb_request *_req)
 	if (!dum->driver)
 		return -ESHUTDOWN;
 
-	local_irq_save(flags);
-	spin_lock(&dum->lock);
+	spin_lock_irqsave(&dum->lock, flags);
 	list_for_each_entry(iter, &ep->queue, queue) {
 		if (&iter->req != _req)
 			continue;
@@ -772,15 +771,16 @@ static int dummy_dequeue(struct usb_ep *_ep, struct usb_request *_req)
 		retval = 0;
 		break;
 	}
-	spin_unlock(&dum->lock);
 
 	if (retval == 0) {
 		dev_dbg(udc_dev(dum),
 				"dequeued req %p from %s, len %d buf %p\n",
 				req, _ep->name, _req->length, _req->buf);
+		spin_unlock(&dum->lock);
 		usb_gadget_giveback_request(_ep, _req);
+		spin_lock(&dum->lock);
 	}
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&dum->lock, flags);
 	return retval;
 }
 
-- 
2.51.0


^ permalink raw reply related	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2025-09-17 13:17 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-09-17  7:44 FAILED: patch "[PATCH] USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels" failed to apply to 5.15-stable tree gregkh
2025-09-17 13:17 ` [PATCH 5.15.y 1/2] usb: gadget: dummy_hcd: remove usage of list iterator past the loop body Sasha Levin
2025-09-17 13:17   ` [PATCH 5.15.y 2/2] USB: gadget: dummy-hcd: Fix locking bug in RT-enabled kernels Sasha Levin

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox