* Forwarded: [PATCH] usb: core: fix memory leak in usb_new_device() error path
2026-04-25 2:12 [syzbot] [usb?] memory leak in hub_event (4) syzbot
@ 2026-04-25 6:36 ` syzbot
2026-04-25 7:52 ` Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure syzbot
` (4 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: syzbot @ 2026-04-25 6:36 UTC (permalink / raw)
To: linux-kernel, syzkaller-bugs
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.
***
Subject: [PATCH] usb: core: fix memory leak in usb_new_device() error path
Author: kartikey406@gmail.com
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
When usb_new_device() fails, it jumps to the 'fail' label which calls
pm_runtime_disable() but never balances the earlier
pm_runtime_get_noresume() call made at the top of the function.
This leaves the PM runtime usage count elevated, preventing
usb_put_dev() in hub_port_connect() from dropping the refcount to zero.
As a result, usb_release_dev() never fires and usb_destroy_configuration()
is never called, leaking all memory allocated during enumeration:
- struct usb_device (2048 bytes) via usb_alloc_dev()
- raw config descriptor (1024 bytes) via usb_get_configuration()
- config metadata ( 8 bytes) via usb_get_configuration()
- interface descriptor ( 64 bytes) via usb_parse_configuration()
- struct device_private ( 256 bytes) via device_private_init()
Fix this by adding pm_runtime_put_noidle() on the fail path to balance
the pm_runtime_get_noresume() at the top of the function.
pm_runtime_put_noidle() is correct here rather than pm_runtime_put()
because we are in a teardown path and must not trigger autosuspend
scheduling.
Reported-by: syzbot+2afd7e71155c7e241560@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=2afd7e71155c7e241560
Signed-off-by: Deepanshu kartikey <kartikey406@gmail.com>
---
drivers/usb/core/hub.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 24960ba9caa9..148fadfbc30b 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2733,6 +2733,7 @@ int usb_new_device(struct usb_device *udev)
usb_set_device_state(udev, USB_STATE_NOTATTACHED);
pm_runtime_disable(&udev->dev);
pm_runtime_set_suspended(&udev->dev);
+ pm_runtime_put_noidle(&udev->dev);
return err;
}
--
2.43.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
2026-04-25 2:12 [syzbot] [usb?] memory leak in hub_event (4) syzbot
2026-04-25 6:36 ` Forwarded: [PATCH] usb: core: fix memory leak in usb_new_device() error path syzbot
@ 2026-04-25 7:52 ` syzbot
2026-04-25 8:58 ` Forwarded: [PATCH] usb: core: hub: fix memory leak in hub_port_connect syzbot
` (3 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: syzbot @ 2026-04-25 7:52 UTC (permalink / raw)
To: linux-kernel, syzkaller-bugs
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.
***
Subject: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
Author: kartikey406@gmail.com
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
create_card() takes a reference on the USB device with usb_get_dev()
and stores the matching usb_put_dev() in card_free(), which is
installed as the snd_card's ->private_free destructor.
However, ->private_free is only assigned near the end of init_card(),
after several failure points (usb_set_interface(), EP type checks,
usb_submit_urb(), the EP1_CMD_GET_DEVICE_INFO exchange, and its
timeout). When any of those fail, init_card() returns an error to
snd_probe(), which calls snd_card_free(card). Because ->private_free
is still NULL, card_free() never runs, the usb_get_dev() reference
is not dropped, and the struct usb_device leaks along with its
descriptor allocations and device_private.
syzbot reproduces this with a malformed UAC3 device whose only valid
altsetting is 0; init_card()'s usb_set_interface(usb_dev, 0, 1) call
fails with -EIO and triggers the leak.
Move the ->private_free assignment into create_card(), immediately
after usb_get_dev(), so that every error path reaching snd_card_free()
balances the reference. card_free()'s callees (snd_usb_caiaq_input_free,
free_urbs, kfree) already tolerate the partially-initialized state
because the chip private area is zero-initialized by snd_card_new().
Reported-by: syzbot+2afd7e71155c7e241560@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=2afd7e71155c7e241560
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
sound/usb/caiaq/device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sound/usb/caiaq/device.c b/sound/usb/caiaq/device.c
index 51177ebfb8c6..f084a0f782ad 100644
--- a/sound/usb/caiaq/device.c
+++ b/sound/usb/caiaq/device.c
@@ -411,6 +411,7 @@ static int create_card(struct usb_device *usb_dev,
cdev = caiaqdev(card);
cdev->chip.dev = usb_get_dev(usb_dev);
+ card->private_free = card_free;
cdev->chip.card = card;
cdev->chip.usb_id = USB_ID(le16_to_cpu(usb_dev->descriptor.idVendor),
le16_to_cpu(usb_dev->descriptor.idProduct));
@@ -500,7 +501,6 @@ static int init_card(struct snd_usb_caiaqdev *cdev)
cdev->vendor_name, cdev->product_name, usbpath);
setup_card(cdev);
- card->private_free = card_free;
return 0;
err_kill_urb:
--
2.43.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Forwarded: [PATCH] usb: core: hub: fix memory leak in hub_port_connect
2026-04-25 2:12 [syzbot] [usb?] memory leak in hub_event (4) syzbot
2026-04-25 6:36 ` Forwarded: [PATCH] usb: core: fix memory leak in usb_new_device() error path syzbot
2026-04-25 7:52 ` Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure syzbot
@ 2026-04-25 8:58 ` syzbot
2026-04-25 9:36 ` syzbot
` (2 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: syzbot @ 2026-04-25 8:58 UTC (permalink / raw)
To: linux-kernel, syzkaller-bugs
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.
***
Subject: [PATCH] usb: core: hub: fix memory leak in hub_port_connect
Author: souradiptodas6@gmail.com
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
master
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 24960ba9caa9..2924ce770066 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -5405,6 +5405,7 @@ static void hub_port_connect(struct usb_hub *hub, int
port1, u16 portstatus,
if (hcd->usb_phy && !hdev->parent)
usb_phy_notify_disconnect(hcd->usb_phy,
udev->speed);
usb_disconnect(&port_dev->child);
+ udev = NULL;
}
/* We can forget about a "removed" device when there's a physical
@@ -5582,6 +5583,7 @@ static void hub_port_connect(struct usb_hub *hub, int
port1, u16 portstatus,
goto loop_disable;
status = hub_power_remaining(hub);
+ udev = NULL;
if (status)
dev_dbg(hub->intfdev, "%dmA power budget left\n",
status);
@@ -5624,6 +5626,8 @@ static void hub_port_connect(struct usb_hub *hub, int
port1, u16 portstatus,
if (status != -ENOTCONN && status != -ENODEV)
hcd->driver->relinquish_port(hcd, port1);
}
+ if (udev)
+ usb_put_dev(udev);
}
/* Handle physical or logical connection change events.
^ permalink raw reply related [flat|nested] 8+ messages in thread* Forwarded: [PATCH] usb: core: hub: fix memory leak in hub_port_connect
2026-04-25 2:12 [syzbot] [usb?] memory leak in hub_event (4) syzbot
` (2 preceding siblings ...)
2026-04-25 8:58 ` Forwarded: [PATCH] usb: core: hub: fix memory leak in hub_port_connect syzbot
@ 2026-04-25 9:36 ` syzbot
2026-04-25 12:54 ` Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure syzbot
2026-04-25 14:43 ` syzbot
5 siblings, 0 replies; 8+ messages in thread
From: syzbot @ 2026-04-25 9:36 UTC (permalink / raw)
To: linux-kernel, syzkaller-bugs
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.
***
Subject: [PATCH] usb: core: hub: fix memory leak in hub_port_connect
Author: souradiptodas6@gmail.com
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
---
drivers/usb/core/hub.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 24960ba9caa9..2924ce770066 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -5405,6 +5405,7 @@ static void hub_port_connect(struct usb_hub *hub, int port1, u16 portstatus,
if (hcd->usb_phy && !hdev->parent)
usb_phy_notify_disconnect(hcd->usb_phy, udev->speed);
usb_disconnect(&port_dev->child);
+ udev = NULL;
}
/* We can forget about a "removed" device when there's a physical
@@ -5582,6 +5583,7 @@ static void hub_port_connect(struct usb_hub *hub, int port1, u16 portstatus,
goto loop_disable;
status = hub_power_remaining(hub);
+ udev = NULL;
if (status)
dev_dbg(hub->intfdev, "%dmA power budget left\n", status);
@@ -5624,6 +5626,8 @@ static void hub_port_connect(struct usb_hub *hub, int port1, u16 portstatus,
if (status != -ENOTCONN && status != -ENODEV)
hcd->driver->relinquish_port(hcd, port1);
}
+ if (udev)
+ usb_put_dev(udev);
}
/* Handle physical or logical connection change events.
--
2.52.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
2026-04-25 2:12 [syzbot] [usb?] memory leak in hub_event (4) syzbot
` (3 preceding siblings ...)
2026-04-25 9:36 ` syzbot
@ 2026-04-25 12:54 ` syzbot
2026-04-25 14:43 ` syzbot
5 siblings, 0 replies; 8+ messages in thread
From: syzbot @ 2026-04-25 12:54 UTC (permalink / raw)
To: linux-kernel, syzkaller-bugs
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.
***
Subject: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
Author: kartikey406@gmail.com
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
create_card() takes a reference on the USB device with usb_get_dev()
and stores the matching usb_put_dev() in card_free(), which is
installed as the snd_card's ->private_free destructor.
However, ->private_free is only assigned near the end of init_card(),
after several failure points (usb_set_interface(), EP type checks,
usb_submit_urb(), the EP1_CMD_GET_DEVICE_INFO exchange, and its
timeout). When any of those fail, init_card() returns an error to
snd_probe(), which calls snd_card_free(card). Because ->private_free
is still NULL, card_free() never runs, the usb_get_dev() reference
is not dropped, and the struct usb_device leaks along with its
descriptor allocations and device_private.
syzbot reproduces this with a malformed UAC3 device whose only valid
altsetting is 0; init_card()'s usb_set_interface(usb_dev, 0, 1) call
fails with -EIO and triggers the leak.
Move the ->private_free assignment into create_card(), immediately
after usb_get_dev(), so that every error path reaching snd_card_free()
balances the reference. card_free()'s callees (snd_usb_caiaq_input_free,
free_urbs, kfree) already tolerate the partially-initialized state
because the chip private area is zero-initialized by snd_card_new().
Reported-by: syzbot+2afd7e71155c7e241560@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=2afd7e71155c7e241560
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
sound/usb/caiaq/device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sound/usb/caiaq/device.c b/sound/usb/caiaq/device.c
index 51177ebfb8c6..f084a0f782ad 100644
--- a/sound/usb/caiaq/device.c
+++ b/sound/usb/caiaq/device.c
@@ -411,6 +411,7 @@ static int create_card(struct usb_device *usb_dev,
cdev = caiaqdev(card);
cdev->chip.dev = usb_get_dev(usb_dev);
+ card->private_free = card_free;
cdev->chip.card = card;
cdev->chip.usb_id = USB_ID(le16_to_cpu(usb_dev->descriptor.idVendor),
le16_to_cpu(usb_dev->descriptor.idProduct));
@@ -500,7 +501,6 @@ static int init_card(struct snd_usb_caiaqdev *cdev)
cdev->vendor_name, cdev->product_name, usbpath);
setup_card(cdev);
- card->private_free = card_free;
return 0;
err_kill_urb:
--
2.43.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
2026-04-25 2:12 [syzbot] [usb?] memory leak in hub_event (4) syzbot
` (4 preceding siblings ...)
2026-04-25 12:54 ` Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure syzbot
@ 2026-04-25 14:43 ` syzbot
2026-04-26 2:33 ` Hillf Danton
5 siblings, 1 reply; 8+ messages in thread
From: syzbot @ 2026-04-25 14:43 UTC (permalink / raw)
To: linux-kernel, syzkaller-bugs
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.
***
Subject: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
Author: kartikey406@gmail.com
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
create_card() takes a reference on the USB device with usb_get_dev()
and stores the matching usb_put_dev() in card_free(), which is
installed as the snd_card's ->private_free destructor.
However, ->private_free is only assigned near the end of init_card(),
after several failure points (usb_set_interface(), EP type checks,
usb_submit_urb(), the EP1_CMD_GET_DEVICE_INFO exchange, and its
timeout). When any of those fail, init_card() returns an error to
snd_probe(), which calls snd_card_free(card). Because ->private_free
is still NULL, card_free() never runs, the usb_get_dev() reference
is not dropped, and the struct usb_device leaks along with its
descriptor allocations and device_private.
syzbot reproduces this with a malformed UAC3 device whose only valid
altsetting is 0; init_card()'s usb_set_interface(usb_dev, 0, 1) call
fails with -EIO and triggers the leak.
Move the ->private_free assignment into create_card(), immediately
after usb_get_dev(), so that every error path reaching snd_card_free()
balances the reference. card_free()'s callees (snd_usb_caiaq_input_free,
free_urbs, kfree) already tolerate the partially-initialized state
because the chip private area is zero-initialized by snd_card_new().
Reported-by: syzbot+2afd7e71155c7e241560@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=2afd7e71155c7e241560
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
sound/usb/caiaq/device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sound/usb/caiaq/device.c b/sound/usb/caiaq/device.c
index 8af0c04041ee..ad9f744b496b 100644
--- a/sound/usb/caiaq/device.c
+++ b/sound/usb/caiaq/device.c
@@ -423,6 +423,7 @@ static int create_card(struct usb_device *usb_dev,
cdev = caiaqdev(card);
cdev->chip.dev = usb_get_dev(usb_dev);
+ card->private_free = card_free;
cdev->chip.card = card;
cdev->chip.usb_id = USB_ID(le16_to_cpu(usb_dev->descriptor.idVendor),
le16_to_cpu(usb_dev->descriptor.idProduct));
@@ -511,7 +512,6 @@ static int init_card(struct snd_usb_caiaqdev *cdev)
scnprintf(card->longname, sizeof(card->longname), "%s %s (%s)",
cdev->vendor_name, cdev->product_name, usbpath);
- card->private_free = card_free;
err = setup_card(cdev);
if (err < 0)
return err;
--
2.43.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Re: Forwarded: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
2026-04-25 14:43 ` syzbot
@ 2026-04-26 2:33 ` Hillf Danton
0 siblings, 0 replies; 8+ messages in thread
From: Hillf Danton @ 2026-04-26 2:33 UTC (permalink / raw)
To: Deepanshu Kartikey; +Cc: syzbot, linux-kernel, syzkaller-bugs
> Date: Sat, 25 Apr 2026 07:43:17 -0700
> For archival purposes, forwarding an incoming command email to
> linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.
>
> ***
>
> Subject: [PATCH] ALSA: caiaq: fix usb_dev refcount leak on probe failure
> Author: kartikey406@gmail.com
>
Deepanshu, can you please correctly fill your Cc list as required to
avoid this forwarded message which is a waste of net bandwidth?
> #syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
>
>
>
> create_card() takes a reference on the USB device with usb_get_dev()
> and stores the matching usb_put_dev() in card_free(), which is
> installed as the snd_card's ->private_free destructor.
>
> However, ->private_free is only assigned near the end of init_card(),
> after several failure points (usb_set_interface(), EP type checks,
> usb_submit_urb(), the EP1_CMD_GET_DEVICE_INFO exchange, and its
> timeout). When any of those fail, init_card() returns an error to
> snd_probe(), which calls snd_card_free(card). Because ->private_free
> is still NULL, card_free() never runs, the usb_get_dev() reference
> is not dropped, and the struct usb_device leaks along with its
> descriptor allocations and device_private.
>
> syzbot reproduces this with a malformed UAC3 device whose only valid
> altsetting is 0; init_card()'s usb_set_interface(usb_dev, 0, 1) call
> fails with -EIO and triggers the leak.
>
> Move the ->private_free assignment into create_card(), immediately
> after usb_get_dev(), so that every error path reaching snd_card_free()
> balances the reference. card_free()'s callees (snd_usb_caiaq_input_free,
> free_urbs, kfree) already tolerate the partially-initialized state
> because the chip private area is zero-initialized by snd_card_new().
>
> Reported-by: syzbot+2afd7e71155c7e241560@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=2afd7e71155c7e241560
> Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
> ---
> sound/usb/caiaq/device.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/sound/usb/caiaq/device.c b/sound/usb/caiaq/device.c
> index 8af0c04041ee..ad9f744b496b 100644
> --- a/sound/usb/caiaq/device.c
> +++ b/sound/usb/caiaq/device.c
> @@ -423,6 +423,7 @@ static int create_card(struct usb_device *usb_dev,
>
> cdev = caiaqdev(card);
> cdev->chip.dev = usb_get_dev(usb_dev);
> + card->private_free = card_free;
> cdev->chip.card = card;
> cdev->chip.usb_id = USB_ID(le16_to_cpu(usb_dev->descriptor.idVendor),
> le16_to_cpu(usb_dev->descriptor.idProduct));
> @@ -511,7 +512,6 @@ static int init_card(struct snd_usb_caiaqdev *cdev)
> scnprintf(card->longname, sizeof(card->longname), "%s %s (%s)",
> cdev->vendor_name, cdev->product_name, usbpath);
>
> - card->private_free = card_free;
> err = setup_card(cdev);
> if (err < 0)
> return err;
> --
> 2.43.0
^ permalink raw reply [flat|nested] 8+ messages in thread