public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Gadget: race between ffs_epfile_io() and ffs_func_eps_disable()
@ 2014-04-08  1:50 Chao Bi
  0 siblings, 0 replies; 2+ messages in thread
From: Chao Bi @ 2014-04-08  1:50 UTC (permalink / raw)
  To: balbi, gregkh; +Cc: linux-usb, linux-kernel


ffs_epfile_io() is called from userspace, while ffs_func_esp_disable() might
be called from USB disconnect interrupt, the two functions would run in parallel
but they are not well protected, that epfile->ep would be removed by
ffs_func_esp_disable() during ffs_epfile_io() is referring this pointer, then it
leads to kernel PANIC.

The scenario is as below:

Thread 1                                 Thread 2
   |                                        |
SyS_read                             dwc3_gadget_disconnect_interrupt
   |                                        |From 772c1c9ef451a9a84e48acf14ba613328920a2f4 Mon Sep 17 00:00:00 2001
From: channing <chao.bi@intel.com>
Date: Wed, 2 Apr 2014 16:37:57 +0800
Subject: [PATCH] Gadget: race between ffs_epfile_io() and
 ffs_func_eps_disable()

ffs_epfile_io() is called from userspace, while ffs_func_esp_disable() might
be called from USB disconnect interrupt, the two functions would run in parallel
but they are not well protected, that epfile->ep would be removed by
ffs_func_esp_disable() during ffs_epfile_io() is referring this pointer, then it
leads to kernel PANIC.

The scenario is as below:

Thread 1                                 Thread 2
   |                                        |
SyS_read                             dwc3_gadget_disconnect_interrupt
   |                                        |
ffs_epfile_read                         reset_config
   |                                        |
ffs_epfile_io                       ffs_func_eps_disable
   |                                        |
 -----                      usb_ep_disable():  epfile->ep->ep->desc = NULL
   |                                        |
usb_ep_align_maybe():                     -----
it refers ep->desc->wMaxPacketSize        -----

Signed-off-by: Chao Bi <chao.bi@intel.com>
---
 drivers/usb/gadget/f_fs.c |    7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c
index 2e164dc..1e12b3e 100644
--- a/drivers/usb/gadget/f_fs.c
+++ b/drivers/usb/gadget/f_fs.c
@@ -745,6 +745,12 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data)
 		 */
 		struct usb_gadget *gadget = epfile->ffs->gadget;
 
+		spin_lock_irq(&epfile->ffs->eps_lock);
+		/* In the meantime, endpoint got disabled or changed. */
+		if (epfile->ep != ep) {
+			spin_unlock_irq(&epfile->ffs->eps_lock);
+			return -ESHUTDOWN;
+		}
 		/*
 		 * Controller may require buffer size to be aligned to
 		 * maxpacketsize of an out endpoint.
@@ -752,6 +758,7 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data)
 		data_len = io_data->read ?
 			   usb_ep_align_maybe(gadget, ep->ep, io_data->len) :
 			   io_data->len;
+		spin_unlock_irq(&epfile->ffs->eps_lock);
 
 		data = kmalloc(data_len, GFP_KERNEL);
 		if (unlikely(!data))
-- 
1.7.9.5


ffs_epfile_read                         reset_config
   |                                        |
ffs_epfile_io                       ffs_func_eps_disable
   |                                        |
 -----                      usb_ep_disable():  epfile->ep->ep->desc = NULL
   |                                        |
usb_ep_align_maybe():                     -----
it refers ep->desc->wMaxPacketSize        -----

Signed-off-by: Chao Bi <chao.bi@intel.com>
---
 drivers/usb/gadget/f_fs.c |    7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c
index 2e164dc..1e12b3e 100644
--- a/drivers/usb/gadget/f_fs.c
+++ b/drivers/usb/gadget/f_fs.c
@@ -745,6 +745,12 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data)
 		 */
 		struct usb_gadget *gadget = epfile->ffs->gadget;
 
+		spin_lock_irq(&epfile->ffs->eps_lock);
+		/* In the meantime, endpoint got disabled or changed. */
+		if (epfile->ep != ep) {
+			spin_unlock_irq(&epfile->ffs->eps_lock);
+			return -ESHUTDOWN;
+		}
 		/*
 		 * Controller may require buffer size to be aligned to
 		 * maxpacketsize of an out endpoint.
@@ -752,6 +758,7 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data)
 		data_len = io_data->read ?
 			   usb_ep_align_maybe(gadget, ep->ep, io_data->len) :
 			   io_data->len;
+		spin_unlock_irq(&epfile->ffs->eps_lock);
 
 		data = kmalloc(data_len, GFP_KERNEL);
 		if (unlikely(!data))
-- 
1.7.9.5




^ permalink raw reply related	[flat|nested] 2+ messages in thread

* [PATCH] Gadget: race between ffs_epfile_io() and ffs_func_eps_disable()
@ 2014-04-08  2:08 Chao Bi
  0 siblings, 0 replies; 2+ messages in thread
From: Chao Bi @ 2014-04-08  2:08 UTC (permalink / raw)
  To: balbi, gregkh; +Cc: linux-usb, linux-kernel


ffs_epfile_io() is called from userspace, while ffs_func_esp_disable() might
be called from USB disconnect interrupt, the two functions would run in parallel
but they are not well protected, that epfile->ep would be removed by
ffs_func_esp_disable() during ffs_epfile_io() is referring this pointer, then it
leads to kernel PANIC.

The scenario is as below:

Thread 1                                 Thread 2
   |                                        |
SyS_read                             dwc3_gadget_disconnect_interrupt
   |                                        |
ffs_epfile_read                         reset_config
   |                                        |
ffs_epfile_io                       ffs_func_eps_disable
   |                                        |
 -----                      usb_ep_disable():  epfile->ep->ep->desc = NULL
   |                                        |
usb_ep_align_maybe():                     -----
it refers ep->desc->wMaxPacketSize        -----

Signed-off-by: Chao Bi <chao.bi@intel.com>
---
 drivers/usb/gadget/f_fs.c |    7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c
index 2e164dc..1e12b3e 100644
--- a/drivers/usb/gadget/f_fs.c
+++ b/drivers/usb/gadget/f_fs.c
@@ -745,6 +745,12 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data)
 		 */
 		struct usb_gadget *gadget = epfile->ffs->gadget;
 
+		spin_lock_irq(&epfile->ffs->eps_lock);
+		/* In the meantime, endpoint got disabled or changed. */
+		if (epfile->ep != ep) {
+			spin_unlock_irq(&epfile->ffs->eps_lock);
+			return -ESHUTDOWN;
+		}
 		/*
 		 * Controller may require buffer size to be aligned to
 		 * maxpacketsize of an out endpoint.
@@ -752,6 +758,7 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data)
 		data_len = io_data->read ?
 			   usb_ep_align_maybe(gadget, ep->ep, io_data->len) :
 			   io_data->len;
+		spin_unlock_irq(&epfile->ffs->eps_lock);
 
 		data = kmalloc(data_len, GFP_KERNEL);
 		if (unlikely(!data))
-- 
1.7.9.5




^ permalink raw reply related	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2014-04-08  1:38 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-04-08  2:08 [PATCH] Gadget: race between ffs_epfile_io() and ffs_func_eps_disable() Chao Bi
  -- strict thread matches above, loose matches on Subject: below --
2014-04-08  1:50 Chao Bi

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