1
0
Fork 0
alistair23-linux/drivers/usb/class/cdc-wdm.c

1182 lines
28 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* cdc-wdm.c
*
* This driver supports USB CDC WCM Device Management.
*
* Copyright (c) 2007-2009 Oliver Neukum
*
* Some code taken from cdc-acm.c
*
* Released under the GPLv2.
*
* Many thanks to Carl Nordbeck
*/
#include <linux/kernel.h>
#include <linux/errno.h>
USB: cdc-wdm: implement IOCTL_WDM_MAX_COMMAND Userspace applications need to know the maximum supported message size. The cdc-wdm driver translates between a character device stream and a message based protocol. Each message is transported as a usb control message with no further encapsulation or syncronization. Each read or write on the character device should translate to exactly one usb control message to ensure that message boundaries are kept intact. That means that the userspace application must know the maximum message size supported by the device and driver, making this size a vital part of the cdc-wdm character device API. CDC WDM and CDC MBIM functions export the maximum supported message size through CDC functional descriptors. The cdc-wdm and cdc_mbim drivers will parse these descriptors and use the value chosen by the device. The only current way for a userspace application to retrive the value is by duplicating the descriptor parsing. This is an unnecessary complex task, and application writers are likely to postpone it, using a fixed value and adding a "todo" item. QMI functions have no way to tell the host what message size they support. The qmi_wwan driver use a fixed value based on protocol recommendations and observed device behaviour. Userspace applications must know and hard code the same value. This scheme will break if we ever encounter a QMI device needing a device specific message size quirk. We are currently unable to support such a device because using a non default size would break the implicit userspace API. The message size is currently a hidden attribute of the cdc-wdm userspace API. Retrieving it is unnecessarily complex, increasing the possibility of drivers and applications using different limits. The resulting errors are hard to debug, and can only be replicated on identical hardware. Exporting the maximum message size from the driver simplifies the task for the userspace application, and creates a unified information source independent of device and function class. It also serves to document that the message size is part of the cdc-wdm userspace API. This proposed API extension has been presented for the authors of userspace applications and libraries using the current API: libmbim, libqmi, uqmi, oFono and ModemManager. The replies were: Aleksander Morgado: "We do really need max message size for MBIM; and as you say, it may be good to have the max message size info also for QMI, so the new ioctl seems a good addition. So +1 from my side, for what it's worth." Dan Williams: "Yeah, +1 here. I'd prefer the sysfs file, but the fact that that doesn't work for fd passing pretty much kills it." No negative replies are so far received. Cc: Aleksander Morgado <aleksander@lanedo.com> Cc: Dan Williams <dcbw@redhat.com> Signed-off-by: Bjørn Mork <bjorn@mork.no> Acked-by: Oliver Neukum <oliver@neukum.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2013-03-17 14:00:06 -06:00
#include <linux/ioctl.h>
#include <linux/slab.h>
#include <linux/module.h>
#include <linux/mutex.h>
#include <linux/uaccess.h>
#include <linux/bitops.h>
#include <linux/poll.h>
#include <linux/usb.h>
#include <linux/usb/cdc.h>
#include <asm/byteorder.h>
#include <asm/unaligned.h>
#include <linux/usb/cdc-wdm.h>
#define DRIVER_AUTHOR "Oliver Neukum"
#define DRIVER_DESC "USB Abstract Control Model driver for USB WCM Device Management"
static const struct usb_device_id wdm_ids[] = {
{
.match_flags = USB_DEVICE_ID_MATCH_INT_CLASS |
USB_DEVICE_ID_MATCH_INT_SUBCLASS,
.bInterfaceClass = USB_CLASS_COMM,
.bInterfaceSubClass = USB_CDC_SUBCLASS_DMM
},
{ }
};
MODULE_DEVICE_TABLE (usb, wdm_ids);
#define WDM_MINOR_BASE 176
#define WDM_IN_USE 1
#define WDM_DISCONNECTING 2
#define WDM_RESULT 3
#define WDM_READ 4
#define WDM_INT_STALL 5
#define WDM_POLL_RUNNING 6
#define WDM_RESPONDING 7
#define WDM_SUSPENDING 8
#define WDM_RESETTING 9
#define WDM_OVERFLOW 10
#define WDM_MAX 16
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
/* we cannot wait forever at flush() */
#define WDM_FLUSH_TIMEOUT (30 * HZ)
/* CDC-WMC r1.1 requires wMaxCommand to be "at least 256 decimal (0x100)" */
#define WDM_DEFAULT_BUFSIZE 256
static DEFINE_MUTEX(wdm_mutex);
static DEFINE_SPINLOCK(wdm_device_list_lock);
static LIST_HEAD(wdm_device_list);
/* --- method tables --- */
struct wdm_device {
u8 *inbuf; /* buffer for response */
u8 *outbuf; /* buffer for command */
u8 *sbuf; /* buffer for status */
u8 *ubuf; /* buffer for copy to user space */
struct urb *command;
struct urb *response;
struct urb *validity;
struct usb_interface *intf;
struct usb_ctrlrequest *orq;
struct usb_ctrlrequest *irq;
spinlock_t iuspin;
unsigned long flags;
u16 bufsize;
u16 wMaxCommand;
u16 wMaxPacketSize;
__le16 inum;
int reslength;
int length;
int read;
int count;
dma_addr_t shandle;
dma_addr_t ihandle;
struct mutex wlock;
struct mutex rlock;
wait_queue_head_t wait;
struct work_struct rxwork;
struct work_struct service_outs_intr;
int werr;
int rerr;
int resp_count;
struct list_head device_list;
int (*manage_power)(struct usb_interface *, int);
};
static struct usb_driver wdm_driver;
/* return intfdata if we own the interface, else look up intf in the list */
static struct wdm_device *wdm_find_device(struct usb_interface *intf)
{
struct wdm_device *desc;
spin_lock(&wdm_device_list_lock);
list_for_each_entry(desc, &wdm_device_list, device_list)
if (desc->intf == intf)
goto found;
desc = NULL;
found:
spin_unlock(&wdm_device_list_lock);
return desc;
}
static struct wdm_device *wdm_find_device_by_minor(int minor)
{
struct wdm_device *desc;
spin_lock(&wdm_device_list_lock);
list_for_each_entry(desc, &wdm_device_list, device_list)
if (desc->intf->minor == minor)
goto found;
desc = NULL;
found:
spin_unlock(&wdm_device_list_lock);
return desc;
}
/* --- callbacks --- */
static void wdm_out_callback(struct urb *urb)
{
struct wdm_device *desc;
unsigned long flags;
desc = urb->context;
spin_lock_irqsave(&desc->iuspin, flags);
desc->werr = urb->status;
spin_unlock_irqrestore(&desc->iuspin, flags);
kfree(desc->outbuf);
desc->outbuf = NULL;
clear_bit(WDM_IN_USE, &desc->flags);
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
wake_up_all(&desc->wait);
}
static void wdm_in_callback(struct urb *urb)
{
unsigned long flags;
struct wdm_device *desc = urb->context;
int status = urb->status;
int length = urb->actual_length;
spin_lock_irqsave(&desc->iuspin, flags);
clear_bit(WDM_RESPONDING, &desc->flags);
if (status) {
switch (status) {
case -ENOENT:
dev_dbg(&desc->intf->dev,
"nonzero urb status received: -ENOENT\n");
goto skip_error;
case -ECONNRESET:
dev_dbg(&desc->intf->dev,
"nonzero urb status received: -ECONNRESET\n");
goto skip_error;
case -ESHUTDOWN:
dev_dbg(&desc->intf->dev,
"nonzero urb status received: -ESHUTDOWN\n");
goto skip_error;
case -EPIPE:
dev_err(&desc->intf->dev,
"nonzero urb status received: -EPIPE\n");
break;
default:
dev_err(&desc->intf->dev,
"Unexpected error %d\n", status);
break;
}
}
/*
* only set a new error if there is no previous error.
* Errors are only cleared during read/open
USB: cdc-wdm: ignore -EPIPE from GetEncapsulatedResponse The driver will forward errors to userspace after turning most of them into -EIO. But all status codes are not equal. The -EPIPE (stall) in particular can be seen more as a result of normal USB signaling than an actual error. The state is automatically cleared by the USB core without intervention from either driver or userspace. And most devices and firmwares will never trigger a stall as a result of GetEncapsulatedResponse. This is in fact a requirement for CDC WDM devices. Quoting from section 7.1 of the CDC WMC spec revision 1.1: The function shall not return STALL in response to GetEncapsulatedResponse. But this driver is also handling GetEncapsulatedResponse on behalf of the qmi_wwan and cdc_mbim drivers. Unfortunately the relevant specs are not as clear wrt stall. So some QMI and MBIM devices *will* occasionally stall, causing the GetEncapsulatedResponse to return an -EPIPE status. Translating this into -EIO for userspace has proven to be harmful. Treating it as an empty read is safer, making the driver behave as if the device was conforming to the CDC WDM spec. There have been numerous reports of issues related to -EPIPE errors from some newer CDC MBIM devices in particular, like for example the Fibocom L831-EAU. Testing on this device has shown that the issues go away if we simply ignore the -EPIPE status. Similar handling of -EPIPE is already known from e.g. usb_get_string() The -EPIPE log message is still kept to let us track devices with this unexpected behaviour, hoping that it attracts attention from firmware developers. Cc: <stable@vger.kernel.org> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100938 Reported-and-tested-by: Christian Ehrig <christian.ehrig@mediamarktsaturn-bt.com> Reported-and-tested-by: Patrick Chilton <chpatrick@gmail.com> Reported-and-tested-by: Andreas Böhler <news@aboehler.at> Signed-off-by: Bjørn Mork <bjorn@mork.no> Acked-by: Oliver Neukum <oneukum@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-09-22 14:18:18 -06:00
* Avoid propagating -EPIPE (stall) to userspace since it is
* better handled as an empty read
*/
USB: cdc-wdm: ignore -EPIPE from GetEncapsulatedResponse The driver will forward errors to userspace after turning most of them into -EIO. But all status codes are not equal. The -EPIPE (stall) in particular can be seen more as a result of normal USB signaling than an actual error. The state is automatically cleared by the USB core without intervention from either driver or userspace. And most devices and firmwares will never trigger a stall as a result of GetEncapsulatedResponse. This is in fact a requirement for CDC WDM devices. Quoting from section 7.1 of the CDC WMC spec revision 1.1: The function shall not return STALL in response to GetEncapsulatedResponse. But this driver is also handling GetEncapsulatedResponse on behalf of the qmi_wwan and cdc_mbim drivers. Unfortunately the relevant specs are not as clear wrt stall. So some QMI and MBIM devices *will* occasionally stall, causing the GetEncapsulatedResponse to return an -EPIPE status. Translating this into -EIO for userspace has proven to be harmful. Treating it as an empty read is safer, making the driver behave as if the device was conforming to the CDC WDM spec. There have been numerous reports of issues related to -EPIPE errors from some newer CDC MBIM devices in particular, like for example the Fibocom L831-EAU. Testing on this device has shown that the issues go away if we simply ignore the -EPIPE status. Similar handling of -EPIPE is already known from e.g. usb_get_string() The -EPIPE log message is still kept to let us track devices with this unexpected behaviour, hoping that it attracts attention from firmware developers. Cc: <stable@vger.kernel.org> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100938 Reported-and-tested-by: Christian Ehrig <christian.ehrig@mediamarktsaturn-bt.com> Reported-and-tested-by: Patrick Chilton <chpatrick@gmail.com> Reported-and-tested-by: Andreas Böhler <news@aboehler.at> Signed-off-by: Bjørn Mork <bjorn@mork.no> Acked-by: Oliver Neukum <oneukum@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-09-22 14:18:18 -06:00
if (desc->rerr == 0 && status != -EPIPE)
desc->rerr = status;
if (length + desc->length > desc->wMaxCommand) {
/* The buffer would overflow */
set_bit(WDM_OVERFLOW, &desc->flags);
} else {
/* we may already be in overflow */
if (!test_bit(WDM_OVERFLOW, &desc->flags)) {
memmove(desc->ubuf + desc->length, desc->inbuf, length);
desc->length += length;
desc->reslength = length;
}
}
skip_error:
if (desc->rerr) {
/*
* Since there was an error, userspace may decide to not read
* any data after poll'ing.
* We should respond to further attempts from the device to send
* data, so that we can get unstuck.
*/
schedule_work(&desc->service_outs_intr);
} else {
set_bit(WDM_READ, &desc->flags);
wake_up(&desc->wait);
}
spin_unlock_irqrestore(&desc->iuspin, flags);
}
static void wdm_int_callback(struct urb *urb)
{
unsigned long flags;
int rv = 0;
int responding;
int status = urb->status;
struct wdm_device *desc;
struct usb_cdc_notification *dr;
desc = urb->context;
dr = (struct usb_cdc_notification *)desc->sbuf;
if (status) {
switch (status) {
case -ESHUTDOWN:
case -ENOENT:
case -ECONNRESET:
return; /* unplug */
case -EPIPE:
set_bit(WDM_INT_STALL, &desc->flags);
dev_err(&desc->intf->dev, "Stall on int endpoint\n");
goto sw; /* halt is cleared in work */
default:
dev_err(&desc->intf->dev,
"nonzero urb status received: %d\n", status);
break;
}
}
if (urb->actual_length < sizeof(struct usb_cdc_notification)) {
dev_err(&desc->intf->dev, "wdm_int_callback - %d bytes\n",
urb->actual_length);
goto exit;
}
switch (dr->bNotificationType) {
case USB_CDC_NOTIFY_RESPONSE_AVAILABLE:
dev_dbg(&desc->intf->dev,
"NOTIFY_RESPONSE_AVAILABLE received: index %d len %d\n",
le16_to_cpu(dr->wIndex), le16_to_cpu(dr->wLength));
break;
case USB_CDC_NOTIFY_NETWORK_CONNECTION:
dev_dbg(&desc->intf->dev,
"NOTIFY_NETWORK_CONNECTION %s network\n",
dr->wValue ? "connected to" : "disconnected from");
goto exit;
case USB_CDC_NOTIFY_SPEED_CHANGE:
dev_dbg(&desc->intf->dev, "SPEED_CHANGE received (len %u)\n",
urb->actual_length);
goto exit;
default:
clear_bit(WDM_POLL_RUNNING, &desc->flags);
dev_err(&desc->intf->dev,
"unknown notification %d received: index %d len %d\n",
dr->bNotificationType,
le16_to_cpu(dr->wIndex),
le16_to_cpu(dr->wLength));
goto exit;
}
spin_lock_irqsave(&desc->iuspin, flags);
responding = test_and_set_bit(WDM_RESPONDING, &desc->flags);
if (!desc->resp_count++ && !responding
&& !test_bit(WDM_DISCONNECTING, &desc->flags)
&& !test_bit(WDM_SUSPENDING, &desc->flags)) {
rv = usb_submit_urb(desc->response, GFP_ATOMIC);
dev_dbg(&desc->intf->dev, "submit response URB %d\n", rv);
}
spin_unlock_irqrestore(&desc->iuspin, flags);
if (rv < 0) {
clear_bit(WDM_RESPONDING, &desc->flags);
if (rv == -EPERM)
return;
if (rv == -ENOMEM) {
sw:
rv = schedule_work(&desc->rxwork);
if (rv)
dev_err(&desc->intf->dev,
"Cannot schedule work\n");
}
}
exit:
rv = usb_submit_urb(urb, GFP_ATOMIC);
if (rv)
dev_err(&desc->intf->dev,
"%s - usb_submit_urb failed with result %d\n",
__func__, rv);
}
static void kill_urbs(struct wdm_device *desc)
{
/* the order here is essential */
usb_kill_urb(desc->command);
usb_kill_urb(desc->validity);
usb_kill_urb(desc->response);
}
static void free_urbs(struct wdm_device *desc)
{
usb_free_urb(desc->validity);
usb_free_urb(desc->response);
usb_free_urb(desc->command);
}
static void cleanup(struct wdm_device *desc)
{
kfree(desc->sbuf);
kfree(desc->inbuf);
kfree(desc->orq);
kfree(desc->irq);
kfree(desc->ubuf);
free_urbs(desc);
kfree(desc);
}
static ssize_t wdm_write
(struct file *file, const char __user *buffer, size_t count, loff_t *ppos)
{
u8 *buf;
int rv = -EMSGSIZE, r, we;
struct wdm_device *desc = file->private_data;
struct usb_ctrlrequest *req;
if (count > desc->wMaxCommand)
count = desc->wMaxCommand;
spin_lock_irq(&desc->iuspin);
we = desc->werr;
desc->werr = 0;
spin_unlock_irq(&desc->iuspin);
if (we < 0)
return usb_translate_errors(we);
buf = memdup_user(buffer, count);
if (IS_ERR(buf))
return PTR_ERR(buf);
/* concurrent writes and disconnect */
r = mutex_lock_interruptible(&desc->wlock);
rv = -ERESTARTSYS;
if (r)
goto out_free_mem;
if (test_bit(WDM_DISCONNECTING, &desc->flags)) {
rv = -ENODEV;
goto out_free_mem_lock;
}
r = usb_autopm_get_interface(desc->intf);
if (r < 0) {
rv = usb_translate_errors(r);
goto out_free_mem_lock;
}
if (!(file->f_flags & O_NONBLOCK))
r = wait_event_interruptible(desc->wait, !test_bit(WDM_IN_USE,
&desc->flags));
else
if (test_bit(WDM_IN_USE, &desc->flags))
r = -EAGAIN;
if (test_bit(WDM_RESETTING, &desc->flags))
r = -EIO;
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
if (test_bit(WDM_DISCONNECTING, &desc->flags))
r = -ENODEV;
if (r < 0) {
rv = r;
goto out_free_mem_pm;
}
req = desc->orq;
usb_fill_control_urb(
desc->command,
interface_to_usbdev(desc->intf),
/* using common endpoint 0 */
usb_sndctrlpipe(interface_to_usbdev(desc->intf), 0),
(unsigned char *)req,
buf,
count,
wdm_out_callback,
desc
);
req->bRequestType = (USB_DIR_OUT | USB_TYPE_CLASS |
USB_RECIP_INTERFACE);
req->bRequest = USB_CDC_SEND_ENCAPSULATED_COMMAND;
req->wValue = 0;
req->wIndex = desc->inum; /* already converted */
req->wLength = cpu_to_le16(count);
set_bit(WDM_IN_USE, &desc->flags);
desc->outbuf = buf;
rv = usb_submit_urb(desc->command, GFP_KERNEL);
if (rv < 0) {
desc->outbuf = NULL;
clear_bit(WDM_IN_USE, &desc->flags);
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
wake_up_all(&desc->wait); /* for wdm_wait_for_response() */
dev_err(&desc->intf->dev, "Tx URB error: %d\n", rv);
rv = usb_translate_errors(rv);
goto out_free_mem_pm;
} else {
dev_dbg(&desc->intf->dev, "Tx URB has been submitted index=%d\n",
le16_to_cpu(req->wIndex));
}
usb_autopm_put_interface(desc->intf);
mutex_unlock(&desc->wlock);
return count;
out_free_mem_pm:
usb_autopm_put_interface(desc->intf);
out_free_mem_lock:
mutex_unlock(&desc->wlock);
out_free_mem:
kfree(buf);
return rv;
}
/*
* Submit the read urb if resp_count is non-zero.
*
* Called with desc->iuspin locked
*/
static int service_outstanding_interrupt(struct wdm_device *desc)
{
int rv = 0;
/* submit read urb only if the device is waiting for it */
if (!desc->resp_count || !--desc->resp_count)
goto out;
USB: cdc-wdm: Fix use after free in service_outstanding_interrupt(). commit 5e5ff0b4b6bcb4d17b7a26ec8bcfc7dd4651684f upstream. syzbot is reporting UAF at usb_submit_urb() [1], for service_outstanding_interrupt() is not checking WDM_DISCONNECTING before calling usb_submit_urb(). Close the race by doing same checks wdm_read() does upon retry. Also, while wdm_read() checks WDM_DISCONNECTING with desc->rlock held, service_interrupt_work() does not hold desc->rlock. Thus, it is possible that usb_submit_urb() is called from service_outstanding_interrupt() from service_interrupt_work() after WDM_DISCONNECTING was set and kill_urbs() from wdm_disconnect() completed. Thus, move kill_urbs() in wdm_disconnect() to after cancel_work_sync() (which makes sure that service_interrupt_work() is no longer running) completed. Although it seems to be safe to dereference desc->intf->dev in service_outstanding_interrupt() even if WDM_DISCONNECTING was already set because desc->rlock or cancel_work_sync() prevents wdm_disconnect() from reaching list_del() before service_outstanding_interrupt() completes, let's not emit error message if WDM_DISCONNECTING is set by wdm_disconnect() while usb_submit_urb() is in progress. [1] https://syzkaller.appspot.com/bug?extid=9e04e2df4a32fb661daf Reported-by: syzbot <syzbot+9e04e2df4a32fb661daf@syzkaller.appspotmail.com> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: stable <stable@vger.kernel.org> Link: https://lore.kernel.org/r/620e2ee0-b9a3-dbda-a25b-a93e0ed03ec5@i-love.sakura.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-12-19 08:25:53 -07:00
if (test_bit(WDM_DISCONNECTING, &desc->flags)) {
rv = -ENODEV;
goto out;
}
if (test_bit(WDM_RESETTING, &desc->flags)) {
rv = -EIO;
goto out;
}
set_bit(WDM_RESPONDING, &desc->flags);
spin_unlock_irq(&desc->iuspin);
rv = usb_submit_urb(desc->response, GFP_KERNEL);
spin_lock_irq(&desc->iuspin);
if (rv) {
USB: cdc-wdm: Fix use after free in service_outstanding_interrupt(). commit 5e5ff0b4b6bcb4d17b7a26ec8bcfc7dd4651684f upstream. syzbot is reporting UAF at usb_submit_urb() [1], for service_outstanding_interrupt() is not checking WDM_DISCONNECTING before calling usb_submit_urb(). Close the race by doing same checks wdm_read() does upon retry. Also, while wdm_read() checks WDM_DISCONNECTING with desc->rlock held, service_interrupt_work() does not hold desc->rlock. Thus, it is possible that usb_submit_urb() is called from service_outstanding_interrupt() from service_interrupt_work() after WDM_DISCONNECTING was set and kill_urbs() from wdm_disconnect() completed. Thus, move kill_urbs() in wdm_disconnect() to after cancel_work_sync() (which makes sure that service_interrupt_work() is no longer running) completed. Although it seems to be safe to dereference desc->intf->dev in service_outstanding_interrupt() even if WDM_DISCONNECTING was already set because desc->rlock or cancel_work_sync() prevents wdm_disconnect() from reaching list_del() before service_outstanding_interrupt() completes, let's not emit error message if WDM_DISCONNECTING is set by wdm_disconnect() while usb_submit_urb() is in progress. [1] https://syzkaller.appspot.com/bug?extid=9e04e2df4a32fb661daf Reported-by: syzbot <syzbot+9e04e2df4a32fb661daf@syzkaller.appspotmail.com> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: stable <stable@vger.kernel.org> Link: https://lore.kernel.org/r/620e2ee0-b9a3-dbda-a25b-a93e0ed03ec5@i-love.sakura.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-12-19 08:25:53 -07:00
if (!test_bit(WDM_DISCONNECTING, &desc->flags))
dev_err(&desc->intf->dev,
"usb_submit_urb failed with result %d\n", rv);
/* make sure the next notification trigger a submit */
clear_bit(WDM_RESPONDING, &desc->flags);
desc->resp_count = 0;
}
out:
return rv;
}
static ssize_t wdm_read
(struct file *file, char __user *buffer, size_t count, loff_t *ppos)
{
int rv, cntr;
int i = 0;
struct wdm_device *desc = file->private_data;
rv = mutex_lock_interruptible(&desc->rlock); /*concurrent reads */
if (rv < 0)
return -ERESTARTSYS;
locking/atomics: COCCINELLE/treewide: Convert trivial ACCESS_ONCE() patterns to READ_ONCE()/WRITE_ONCE() Please do not apply this to mainline directly, instead please re-run the coccinelle script shown below and apply its output. For several reasons, it is desirable to use {READ,WRITE}_ONCE() in preference to ACCESS_ONCE(), and new code is expected to use one of the former. So far, there's been no reason to change most existing uses of ACCESS_ONCE(), as these aren't harmful, and changing them results in churn. However, for some features, the read/write distinction is critical to correct operation. To distinguish these cases, separate read/write accessors must be used. This patch migrates (most) remaining ACCESS_ONCE() instances to {READ,WRITE}_ONCE(), using the following coccinelle script: ---- // Convert trivial ACCESS_ONCE() uses to equivalent READ_ONCE() and // WRITE_ONCE() // $ make coccicheck COCCI=/home/mark/once.cocci SPFLAGS="--include-headers" MODE=patch virtual patch @ depends on patch @ expression E1, E2; @@ - ACCESS_ONCE(E1) = E2 + WRITE_ONCE(E1, E2) @ depends on patch @ expression E; @@ - ACCESS_ONCE(E) + READ_ONCE(E) ---- Signed-off-by: Mark Rutland <mark.rutland@arm.com> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: davem@davemloft.net Cc: linux-arch@vger.kernel.org Cc: mpe@ellerman.id.au Cc: shuah@kernel.org Cc: snitzer@redhat.com Cc: thor.thayer@linux.intel.com Cc: tj@kernel.org Cc: viro@zeniv.linux.org.uk Cc: will.deacon@arm.com Link: http://lkml.kernel.org/r/1508792849-3115-19-git-send-email-paulmck@linux.vnet.ibm.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-10-23 15:07:29 -06:00
cntr = READ_ONCE(desc->length);
if (cntr == 0) {
desc->read = 0;
retry:
if (test_bit(WDM_DISCONNECTING, &desc->flags)) {
rv = -ENODEV;
goto err;
}
if (test_bit(WDM_OVERFLOW, &desc->flags)) {
clear_bit(WDM_OVERFLOW, &desc->flags);
rv = -ENOBUFS;
goto err;
}
i++;
if (file->f_flags & O_NONBLOCK) {
if (!test_bit(WDM_READ, &desc->flags)) {
rv = -EAGAIN;
goto err;
}
rv = 0;
} else {
rv = wait_event_interruptible(desc->wait,
test_bit(WDM_READ, &desc->flags));
}
/* may have happened while we slept */
if (test_bit(WDM_DISCONNECTING, &desc->flags)) {
rv = -ENODEV;
goto err;
}
if (test_bit(WDM_RESETTING, &desc->flags)) {
rv = -EIO;
goto err;
}
usb_mark_last_busy(interface_to_usbdev(desc->intf));
if (rv < 0) {
rv = -ERESTARTSYS;
goto err;
}
spin_lock_irq(&desc->iuspin);
if (desc->rerr) { /* read completed, error happened */
rv = usb_translate_errors(desc->rerr);
desc->rerr = 0;
spin_unlock_irq(&desc->iuspin);
goto err;
}
/*
* recheck whether we've lost the race
* against the completion handler
*/
if (!test_bit(WDM_READ, &desc->flags)) { /* lost race */
spin_unlock_irq(&desc->iuspin);
goto retry;
}
if (!desc->reslength) { /* zero length read */
dev_dbg(&desc->intf->dev, "zero length - clearing WDM_READ\n");
clear_bit(WDM_READ, &desc->flags);
rv = service_outstanding_interrupt(desc);
spin_unlock_irq(&desc->iuspin);
if (rv < 0)
goto err;
goto retry;
}
cntr = desc->length;
spin_unlock_irq(&desc->iuspin);
}
if (cntr > count)
cntr = count;
rv = copy_to_user(buffer, desc->ubuf, cntr);
if (rv > 0) {
rv = -EFAULT;
goto err;
}
spin_lock_irq(&desc->iuspin);
for (i = 0; i < desc->length - cntr; i++)
desc->ubuf[i] = desc->ubuf[i + cntr];
desc->length -= cntr;
/* in case we had outstanding data */
if (!desc->length) {
clear_bit(WDM_READ, &desc->flags);
service_outstanding_interrupt(desc);
}
spin_unlock_irq(&desc->iuspin);
rv = cntr;
err:
mutex_unlock(&desc->rlock);
return rv;
}
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
static int wdm_wait_for_response(struct file *file, long timeout)
{
struct wdm_device *desc = file->private_data;
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
long rv; /* Use long here because (int) MAX_SCHEDULE_TIMEOUT < 0. */
/*
* Needs both flags. We cannot do with one because resetting it would
* cause a race with write() yet we need to signal a disconnect.
*/
rv = wait_event_interruptible_timeout(desc->wait,
!test_bit(WDM_IN_USE, &desc->flags) ||
test_bit(WDM_DISCONNECTING, &desc->flags),
timeout);
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
/*
* To report the correct error. This is best effort.
* We are inevitably racing with the hardware.
*/
if (test_bit(WDM_DISCONNECTING, &desc->flags))
return -ENODEV;
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
if (!rv)
return -EIO;
if (rv < 0)
return -EINTR;
spin_lock_irq(&desc->iuspin);
rv = desc->werr;
desc->werr = 0;
spin_unlock_irq(&desc->iuspin);
return usb_translate_errors(rv);
}
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
/*
* You need to send a signal when you react to malicious or defective hardware.
* Also, don't abort when fsync() returned -EINVAL, for older kernels which do
* not implement wdm_flush() will return -EINVAL.
*/
static int wdm_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
return wdm_wait_for_response(file, MAX_SCHEDULE_TIMEOUT);
}
/*
* Same with wdm_fsync(), except it uses finite timeout in order to react to
* malicious or defective hardware which ceased communication after close() was
* implicitly called due to process termination.
*/
static int wdm_flush(struct file *file, fl_owner_t id)
{
return wdm_wait_for_response(file, WDM_FLUSH_TIMEOUT);
}
static __poll_t wdm_poll(struct file *file, struct poll_table_struct *wait)
{
struct wdm_device *desc = file->private_data;
unsigned long flags;
__poll_t mask = 0;
spin_lock_irqsave(&desc->iuspin, flags);
if (test_bit(WDM_DISCONNECTING, &desc->flags)) {
mask = EPOLLHUP | EPOLLERR;
spin_unlock_irqrestore(&desc->iuspin, flags);
goto desc_out;
}
if (test_bit(WDM_READ, &desc->flags))
mask = EPOLLIN | EPOLLRDNORM;
if (desc->rerr || desc->werr)
mask |= EPOLLERR;
if (!test_bit(WDM_IN_USE, &desc->flags))
mask |= EPOLLOUT | EPOLLWRNORM;
spin_unlock_irqrestore(&desc->iuspin, flags);
poll_wait(file, &desc->wait, wait);
desc_out:
return mask;
}
static int wdm_open(struct inode *inode, struct file *file)
{
int minor = iminor(inode);
int rv = -ENODEV;
struct usb_interface *intf;
struct wdm_device *desc;
mutex_lock(&wdm_mutex);
desc = wdm_find_device_by_minor(minor);
if (!desc)
goto out;
intf = desc->intf;
if (test_bit(WDM_DISCONNECTING, &desc->flags))
goto out;
file->private_data = desc;
rv = usb_autopm_get_interface(desc->intf);
if (rv < 0) {
dev_err(&desc->intf->dev, "Error autopm - %d\n", rv);
goto out;
}
/* using write lock to protect desc->count */
mutex_lock(&desc->wlock);
if (!desc->count++) {
desc->werr = 0;
desc->rerr = 0;
rv = usb_submit_urb(desc->validity, GFP_KERNEL);
if (rv < 0) {
desc->count--;
dev_err(&desc->intf->dev,
"Error submitting int urb - %d\n", rv);
rv = usb_translate_errors(rv);
}
} else {
rv = 0;
}
mutex_unlock(&desc->wlock);
if (desc->count == 1)
desc->manage_power(intf, 1);
usb_autopm_put_interface(desc->intf);
out:
mutex_unlock(&wdm_mutex);
return rv;
}
static int wdm_release(struct inode *inode, struct file *file)
{
struct wdm_device *desc = file->private_data;
mutex_lock(&wdm_mutex);
/* using write lock to protect desc->count */
mutex_lock(&desc->wlock);
desc->count--;
mutex_unlock(&desc->wlock);
if (!desc->count) {
if (!test_bit(WDM_DISCONNECTING, &desc->flags)) {
dev_dbg(&desc->intf->dev, "wdm_release: cleanup\n");
kill_urbs(desc);
spin_lock_irq(&desc->iuspin);
desc->resp_count = 0;
spin_unlock_irq(&desc->iuspin);
desc->manage_power(desc->intf, 0);
} else {
/* must avoid dev_printk here as desc->intf is invalid */
pr_debug(KBUILD_MODNAME " %s: device gone - cleaning up\n", __func__);
cleanup(desc);
}
}
mutex_unlock(&wdm_mutex);
return 0;
}
USB: cdc-wdm: implement IOCTL_WDM_MAX_COMMAND Userspace applications need to know the maximum supported message size. The cdc-wdm driver translates between a character device stream and a message based protocol. Each message is transported as a usb control message with no further encapsulation or syncronization. Each read or write on the character device should translate to exactly one usb control message to ensure that message boundaries are kept intact. That means that the userspace application must know the maximum message size supported by the device and driver, making this size a vital part of the cdc-wdm character device API. CDC WDM and CDC MBIM functions export the maximum supported message size through CDC functional descriptors. The cdc-wdm and cdc_mbim drivers will parse these descriptors and use the value chosen by the device. The only current way for a userspace application to retrive the value is by duplicating the descriptor parsing. This is an unnecessary complex task, and application writers are likely to postpone it, using a fixed value and adding a "todo" item. QMI functions have no way to tell the host what message size they support. The qmi_wwan driver use a fixed value based on protocol recommendations and observed device behaviour. Userspace applications must know and hard code the same value. This scheme will break if we ever encounter a QMI device needing a device specific message size quirk. We are currently unable to support such a device because using a non default size would break the implicit userspace API. The message size is currently a hidden attribute of the cdc-wdm userspace API. Retrieving it is unnecessarily complex, increasing the possibility of drivers and applications using different limits. The resulting errors are hard to debug, and can only be replicated on identical hardware. Exporting the maximum message size from the driver simplifies the task for the userspace application, and creates a unified information source independent of device and function class. It also serves to document that the message size is part of the cdc-wdm userspace API. This proposed API extension has been presented for the authors of userspace applications and libraries using the current API: libmbim, libqmi, uqmi, oFono and ModemManager. The replies were: Aleksander Morgado: "We do really need max message size for MBIM; and as you say, it may be good to have the max message size info also for QMI, so the new ioctl seems a good addition. So +1 from my side, for what it's worth." Dan Williams: "Yeah, +1 here. I'd prefer the sysfs file, but the fact that that doesn't work for fd passing pretty much kills it." No negative replies are so far received. Cc: Aleksander Morgado <aleksander@lanedo.com> Cc: Dan Williams <dcbw@redhat.com> Signed-off-by: Bjørn Mork <bjorn@mork.no> Acked-by: Oliver Neukum <oliver@neukum.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2013-03-17 14:00:06 -06:00
static long wdm_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
struct wdm_device *desc = file->private_data;
int rv = 0;
switch (cmd) {
case IOCTL_WDM_MAX_COMMAND:
if (copy_to_user((void __user *)arg, &desc->wMaxCommand, sizeof(desc->wMaxCommand)))
rv = -EFAULT;
break;
default:
rv = -ENOTTY;
}
return rv;
}
static const struct file_operations wdm_fops = {
.owner = THIS_MODULE,
.read = wdm_read,
.write = wdm_write,
USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). commit 37d2a36394d954413a495da61da1b2a51ecd28ab upstream. syzbot is reporting hung task at wdm_flush() [1], for there is a circular dependency that wdm_flush() from flip_close() for /dev/cdc-wdm0 forever waits for /dev/raw-gadget to be closed while close() for /dev/raw-gadget cannot be called unless close() for /dev/cdc-wdm0 completes. Tetsuo Handa considered that such circular dependency is an usage error [2] which corresponds to an unresponding broken hardware [3]. But Alan Stern responded that we should be prepared for such hardware [4]. Therefore, this patch changes wdm_flush() to use wait_event_interruptible_timeout() which gives up after 30 seconds, for hardware that remains silent must be ignored. The 30 seconds are coming out of thin air. Changing wait_event() to wait_event_interruptible_timeout() makes error reporting from close() syscall less reliable. To compensate it, this patch also implements wdm_fsync() which does not use timeout. Those who want to be very sure that data has gone out to the device are now advised to call fsync(), with a caveat that fsync() can return -EINVAL when running on older kernels which do not implement wdm_fsync(). This patch also fixes three more problems (listed below) found during exhaustive discussion and testing. Since multiple threads can concurrently call wdm_write()/wdm_flush(), we need to use wake_up_all() whenever clearing WDM_IN_USE in order to make sure that all waiters are woken up. Also, error reporting needs to use fetch-and-clear approach in order not to report same error for multiple times. Since wdm_flush() checks WDM_DISCONNECTING, wdm_write() should as well check WDM_DISCONNECTING. In wdm_flush(), since locks are not held, it is not safe to dereference desc->intf after checking that WDM_DISCONNECTING is not set [5]. Thus, remove dev_err() from wdm_flush(). [1] https://syzkaller.appspot.com/bug?id=e7b761593b23eb50855b9ea31e3be5472b711186 [2] https://lkml.kernel.org/r/27b7545e-8f41-10b8-7c02-e35a08eb1611@i-love.sakura.ne.jp [3] https://lkml.kernel.org/r/79ba410f-e0ef-2465-b94f-6b9a4a82adf5@i-love.sakura.ne.jp [4] https://lkml.kernel.org/r/20200530011040.GB12419@rowland.harvard.edu [5] https://lkml.kernel.org/r/c85331fc-874c-6e46-a77f-0ef1dc075308@i-love.sakura.ne.jp Reported-by: syzbot <syzbot+854768b99f19e89d7f81@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> Co-developed-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Oliver Neukum <oneukum@suse.com> Cc: Alan Stern <stern@rowland.harvard.edu> Link: https://lore.kernel.org/r/20200928141755.3476-1-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-09-28 08:17:55 -06:00
.fsync = wdm_fsync,
.open = wdm_open,
.flush = wdm_flush,
.release = wdm_release,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-15 10:52:59 -06:00
.poll = wdm_poll,
USB: cdc-wdm: implement IOCTL_WDM_MAX_COMMAND Userspace applications need to know the maximum supported message size. The cdc-wdm driver translates between a character device stream and a message based protocol. Each message is transported as a usb control message with no further encapsulation or syncronization. Each read or write on the character device should translate to exactly one usb control message to ensure that message boundaries are kept intact. That means that the userspace application must know the maximum message size supported by the device and driver, making this size a vital part of the cdc-wdm character device API. CDC WDM and CDC MBIM functions export the maximum supported message size through CDC functional descriptors. The cdc-wdm and cdc_mbim drivers will parse these descriptors and use the value chosen by the device. The only current way for a userspace application to retrive the value is by duplicating the descriptor parsing. This is an unnecessary complex task, and application writers are likely to postpone it, using a fixed value and adding a "todo" item. QMI functions have no way to tell the host what message size they support. The qmi_wwan driver use a fixed value based on protocol recommendations and observed device behaviour. Userspace applications must know and hard code the same value. This scheme will break if we ever encounter a QMI device needing a device specific message size quirk. We are currently unable to support such a device because using a non default size would break the implicit userspace API. The message size is currently a hidden attribute of the cdc-wdm userspace API. Retrieving it is unnecessarily complex, increasing the possibility of drivers and applications using different limits. The resulting errors are hard to debug, and can only be replicated on identical hardware. Exporting the maximum message size from the driver simplifies the task for the userspace application, and creates a unified information source independent of device and function class. It also serves to document that the message size is part of the cdc-wdm userspace API. This proposed API extension has been presented for the authors of userspace applications and libraries using the current API: libmbim, libqmi, uqmi, oFono and ModemManager. The replies were: Aleksander Morgado: "We do really need max message size for MBIM; and as you say, it may be good to have the max message size info also for QMI, so the new ioctl seems a good addition. So +1 from my side, for what it's worth." Dan Williams: "Yeah, +1 here. I'd prefer the sysfs file, but the fact that that doesn't work for fd passing pretty much kills it." No negative replies are so far received. Cc: Aleksander Morgado <aleksander@lanedo.com> Cc: Dan Williams <dcbw@redhat.com> Signed-off-by: Bjørn Mork <bjorn@mork.no> Acked-by: Oliver Neukum <oliver@neukum.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2013-03-17 14:00:06 -06:00
.unlocked_ioctl = wdm_ioctl,
.compat_ioctl = wdm_ioctl,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-15 10:52:59 -06:00
.llseek = noop_llseek,
};
static struct usb_class_driver wdm_class = {
.name = "cdc-wdm%d",
.fops = &wdm_fops,
.minor_base = WDM_MINOR_BASE,
};
/* --- error handling --- */
static void wdm_rxwork(struct work_struct *work)
{
struct wdm_device *desc = container_of(work, struct wdm_device, rxwork);
unsigned long flags;
int rv = 0;
int responding;
spin_lock_irqsave(&desc->iuspin, flags);
if (test_bit(WDM_DISCONNECTING, &desc->flags)) {
spin_unlock_irqrestore(&desc->iuspin, flags);
} else {
responding = test_and_set_bit(WDM_RESPONDING, &desc->flags);
spin_unlock_irqrestore(&desc->iuspin, flags);
if (!responding)
rv = usb_submit_urb(desc->response, GFP_KERNEL);
if (rv < 0 && rv != -EPERM) {
spin_lock_irqsave(&desc->iuspin, flags);
clear_bit(WDM_RESPONDING, &desc->flags);
if (!test_bit(WDM_DISCONNECTING, &desc->flags))
schedule_work(&desc->rxwork);
spin_unlock_irqrestore(&desc->iuspin, flags);
}
}
}
static void service_interrupt_work(struct work_struct *work)
{
struct wdm_device *desc;
desc = container_of(work, struct wdm_device, service_outs_intr);
spin_lock_irq(&desc->iuspin);
service_outstanding_interrupt(desc);
if (!desc->resp_count) {
set_bit(WDM_READ, &desc->flags);
wake_up(&desc->wait);
}
spin_unlock_irq(&desc->iuspin);
}
/* --- hotplug --- */
static int wdm_create(struct usb_interface *intf, struct usb_endpoint_descriptor *ep,
u16 bufsize, int (*manage_power)(struct usb_interface *, int))
{
int rv = -ENOMEM;
struct wdm_device *desc;
desc = kzalloc(sizeof(struct wdm_device), GFP_KERNEL);
if (!desc)
goto out;
INIT_LIST_HEAD(&desc->device_list);
mutex_init(&desc->rlock);
mutex_init(&desc->wlock);
spin_lock_init(&desc->iuspin);
init_waitqueue_head(&desc->wait);
desc->wMaxCommand = bufsize;
/* this will be expanded and needed in hardware endianness */
desc->inum = cpu_to_le16((u16)intf->cur_altsetting->desc.bInterfaceNumber);
desc->intf = intf;
INIT_WORK(&desc->rxwork, wdm_rxwork);
INIT_WORK(&desc->service_outs_intr, service_interrupt_work);
rv = -EINVAL;
if (!usb_endpoint_is_int_in(ep))
goto err;
USB: use usb_endpoint_maxp() instead of le16_to_cpu() Now ${LINUX}/drivers/usb/* can use usb_endpoint_maxp(desc) to get maximum packet size instead of le16_to_cpu(desc->wMaxPacketSize). This patch fix it up Cc: Armin Fuerst <fuerst@in.tum.de> Cc: Pavel Machek <pavel@ucw.cz> Cc: Johannes Erdfelt <johannes@erdfelt.com> Cc: Vojtech Pavlik <vojtech@suse.cz> Cc: Oliver Neukum <oliver@neukum.name> Cc: David Kubicek <dave@awk.cz> Cc: Johan Hovold <jhovold@gmail.com> Cc: Brad Hards <bhards@bigpond.net.au> Acked-by: Felipe Balbi <balbi@ti.com> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Thomas Dahlmann <dahlmann.thomas@arcor.de> Cc: David Brownell <david-b@pacbell.net> Cc: David Lopo <dlopo@chipidea.mips.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Michal Nazarewicz <m.nazarewicz@samsung.com> Cc: Xie Xiaobo <X.Xie@freescale.com> Cc: Li Yang <leoli@freescale.com> Cc: Jiang Bo <tanya.jiang@freescale.com> Cc: Yuan-hsin Chen <yhchen@faraday-tech.com> Cc: Darius Augulis <augulis.darius@gmail.com> Cc: Xiaochen Shen <xiaochen.shen@intel.com> Cc: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Cc: OKI SEMICONDUCTOR, <toshiharu-linux@dsn.okisemi.com> Cc: Robert Jarzmik <robert.jarzmik@free.fr> Cc: Ben Dooks <ben@simtec.co.uk> Cc: Thomas Abraham <thomas.ab@samsung.com> Cc: Herbert Pötzl <herbert@13thfloor.at> Cc: Arnaud Patard <arnaud.patard@rtp-net.org> Cc: Roman Weissgaerber <weissg@vienna.at> Acked-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Cc: Tony Olech <tony.olech@elandigitalsystems.com> Cc: Florian Floe Echtler <echtler@fs.tum.de> Cc: Christian Lucht <lucht@codemercs.com> Cc: Juergen Stuber <starblue@sourceforge.net> Cc: Georges Toth <g.toth@e-biz.lu> Cc: Bill Ryder <bryder@sgi.com> Cc: Kuba Ober <kuba@mareimbrium.org> Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com> Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-08-23 04:12:03 -06:00
desc->wMaxPacketSize = usb_endpoint_maxp(ep);
desc->orq = kmalloc(sizeof(struct usb_ctrlrequest), GFP_KERNEL);
if (!desc->orq)
goto err;
desc->irq = kmalloc(sizeof(struct usb_ctrlrequest), GFP_KERNEL);
if (!desc->irq)
goto err;
desc->validity = usb_alloc_urb(0, GFP_KERNEL);
if (!desc->validity)
goto err;
desc->response = usb_alloc_urb(0, GFP_KERNEL);
if (!desc->response)
goto err;
desc->command = usb_alloc_urb(0, GFP_KERNEL);
if (!desc->command)
goto err;
desc->ubuf = kmalloc(desc->wMaxCommand, GFP_KERNEL);
if (!desc->ubuf)
goto err;
desc->sbuf = kmalloc(desc->wMaxPacketSize, GFP_KERNEL);
if (!desc->sbuf)
goto err;
desc->inbuf = kmalloc(desc->wMaxCommand, GFP_KERNEL);
if (!desc->inbuf)
goto err;
usb_fill_int_urb(
desc->validity,
interface_to_usbdev(intf),
usb_rcvintpipe(interface_to_usbdev(intf), ep->bEndpointAddress),
desc->sbuf,
desc->wMaxPacketSize,
wdm_int_callback,
desc,
ep->bInterval
);
desc->irq->bRequestType = (USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE);
desc->irq->bRequest = USB_CDC_GET_ENCAPSULATED_RESPONSE;
desc->irq->wValue = 0;
desc->irq->wIndex = desc->inum; /* already converted */
desc->irq->wLength = cpu_to_le16(desc->wMaxCommand);
usb_fill_control_urb(
desc->response,
interface_to_usbdev(intf),
/* using common endpoint 0 */
usb_rcvctrlpipe(interface_to_usbdev(desc->intf), 0),
(unsigned char *)desc->irq,
desc->inbuf,
desc->wMaxCommand,
wdm_in_callback,
desc
);
desc->manage_power = manage_power;
spin_lock(&wdm_device_list_lock);
list_add(&desc->device_list, &wdm_device_list);
spin_unlock(&wdm_device_list_lock);
rv = usb_register_dev(intf, &wdm_class);
if (rv < 0)
goto err;
else
dev_info(&intf->dev, "%s: USB WDM device\n", dev_name(intf->usb_dev));
out:
return rv;
err:
spin_lock(&wdm_device_list_lock);
list_del(&desc->device_list);
spin_unlock(&wdm_device_list_lock);
cleanup(desc);
return rv;
}
static int wdm_manage_power(struct usb_interface *intf, int on)
{
/* need autopm_get/put here to ensure the usbcore sees the new value */
int rv = usb_autopm_get_interface(intf);
intf->needs_remote_wakeup = on;
if (!rv)
usb_autopm_put_interface(intf);
return 0;
}
static int wdm_probe(struct usb_interface *intf, const struct usb_device_id *id)
{
int rv = -EINVAL;
struct usb_host_interface *iface;
struct usb_endpoint_descriptor *ep;
struct usb_cdc_parsed_header hdr;
u8 *buffer = intf->altsetting->extra;
int buflen = intf->altsetting->extralen;
u16 maxcom = WDM_DEFAULT_BUFSIZE;
if (!buffer)
goto err;
cdc_parse_cdc_header(&hdr, intf, buffer, buflen);
if (hdr.usb_cdc_dmm_desc)
maxcom = le16_to_cpu(hdr.usb_cdc_dmm_desc->wMaxCommand);
iface = intf->cur_altsetting;
if (iface->desc.bNumEndpoints != 1)
goto err;
ep = &iface->endpoint[0].desc;
rv = wdm_create(intf, ep, maxcom, &wdm_manage_power);
err:
return rv;
}
/**
* usb_cdc_wdm_register - register a WDM subdriver
* @intf: usb interface the subdriver will associate with
* @ep: interrupt endpoint to monitor for notifications
* @bufsize: maximum message size to support for read/write
*
* Create WDM usb class character device and associate it with intf
* without binding, allowing another driver to manage the interface.
*
* The subdriver will manage the given interrupt endpoint exclusively
* and will issue control requests referring to the given intf. It
* will otherwise avoid interferring, and in particular not do
* usb_set_intfdata/usb_get_intfdata on intf.
*
* The return value is a pointer to the subdriver's struct usb_driver.
* The registering driver is responsible for calling this subdriver's
* disconnect, suspend, resume, pre_reset and post_reset methods from
* its own.
*/
struct usb_driver *usb_cdc_wdm_register(struct usb_interface *intf,
struct usb_endpoint_descriptor *ep,
int bufsize,
int (*manage_power)(struct usb_interface *, int))
{
int rv;
rv = wdm_create(intf, ep, bufsize, manage_power);
if (rv < 0)
goto err;
return &wdm_driver;
err:
return ERR_PTR(rv);
}
EXPORT_SYMBOL(usb_cdc_wdm_register);
static void wdm_disconnect(struct usb_interface *intf)
{
struct wdm_device *desc;
unsigned long flags;
usb_deregister_dev(intf, &wdm_class);
desc = wdm_find_device(intf);
mutex_lock(&wdm_mutex);
/* the spinlock makes sure no new urbs are generated in the callbacks */
spin_lock_irqsave(&desc->iuspin, flags);
set_bit(WDM_DISCONNECTING, &desc->flags);
set_bit(WDM_READ, &desc->flags);
spin_unlock_irqrestore(&desc->iuspin, flags);
wake_up_all(&desc->wait);
mutex_lock(&desc->rlock);
mutex_lock(&desc->wlock);
cancel_work_sync(&desc->rxwork);
cancel_work_sync(&desc->service_outs_intr);
USB: cdc-wdm: Fix use after free in service_outstanding_interrupt(). commit 5e5ff0b4b6bcb4d17b7a26ec8bcfc7dd4651684f upstream. syzbot is reporting UAF at usb_submit_urb() [1], for service_outstanding_interrupt() is not checking WDM_DISCONNECTING before calling usb_submit_urb(). Close the race by doing same checks wdm_read() does upon retry. Also, while wdm_read() checks WDM_DISCONNECTING with desc->rlock held, service_interrupt_work() does not hold desc->rlock. Thus, it is possible that usb_submit_urb() is called from service_outstanding_interrupt() from service_interrupt_work() after WDM_DISCONNECTING was set and kill_urbs() from wdm_disconnect() completed. Thus, move kill_urbs() in wdm_disconnect() to after cancel_work_sync() (which makes sure that service_interrupt_work() is no longer running) completed. Although it seems to be safe to dereference desc->intf->dev in service_outstanding_interrupt() even if WDM_DISCONNECTING was already set because desc->rlock or cancel_work_sync() prevents wdm_disconnect() from reaching list_del() before service_outstanding_interrupt() completes, let's not emit error message if WDM_DISCONNECTING is set by wdm_disconnect() while usb_submit_urb() is in progress. [1] https://syzkaller.appspot.com/bug?extid=9e04e2df4a32fb661daf Reported-by: syzbot <syzbot+9e04e2df4a32fb661daf@syzkaller.appspotmail.com> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: stable <stable@vger.kernel.org> Link: https://lore.kernel.org/r/620e2ee0-b9a3-dbda-a25b-a93e0ed03ec5@i-love.sakura.ne.jp Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-12-19 08:25:53 -07:00
kill_urbs(desc);
mutex_unlock(&desc->wlock);
mutex_unlock(&desc->rlock);
/* the desc->intf pointer used as list key is now invalid */
spin_lock(&wdm_device_list_lock);
list_del(&desc->device_list);
spin_unlock(&wdm_device_list_lock);
if (!desc->count)
cleanup(desc);
else
dev_dbg(&intf->dev, "%d open files - postponing cleanup\n", desc->count);
mutex_unlock(&wdm_mutex);
}
#ifdef CONFIG_PM
static int wdm_suspend(struct usb_interface *intf, pm_message_t message)
{
struct wdm_device *desc = wdm_find_device(intf);
int rv = 0;
dev_dbg(&desc->intf->dev, "wdm%d_suspend\n", intf->minor);
/* if this is an autosuspend the caller does the locking */
if (!PMSG_IS_AUTO(message)) {
mutex_lock(&desc->rlock);
mutex_lock(&desc->wlock);
}
spin_lock_irq(&desc->iuspin);
if (PMSG_IS_AUTO(message) &&
(test_bit(WDM_IN_USE, &desc->flags)
|| test_bit(WDM_RESPONDING, &desc->flags))) {
spin_unlock_irq(&desc->iuspin);
rv = -EBUSY;
} else {
set_bit(WDM_SUSPENDING, &desc->flags);
spin_unlock_irq(&desc->iuspin);
/* callback submits work - order is essential */
kill_urbs(desc);
cancel_work_sync(&desc->rxwork);
cancel_work_sync(&desc->service_outs_intr);
}
if (!PMSG_IS_AUTO(message)) {
mutex_unlock(&desc->wlock);
mutex_unlock(&desc->rlock);
}
return rv;
}
#endif
static int recover_from_urb_loss(struct wdm_device *desc)
{
int rv = 0;
if (desc->count) {
rv = usb_submit_urb(desc->validity, GFP_NOIO);
if (rv < 0)
dev_err(&desc->intf->dev,
"Error resume submitting int urb - %d\n", rv);
}
return rv;
}
#ifdef CONFIG_PM
static int wdm_resume(struct usb_interface *intf)
{
struct wdm_device *desc = wdm_find_device(intf);
int rv;
dev_dbg(&desc->intf->dev, "wdm%d_resume\n", intf->minor);
clear_bit(WDM_SUSPENDING, &desc->flags);
rv = recover_from_urb_loss(desc);
return rv;
}
#endif
static int wdm_pre_reset(struct usb_interface *intf)
{
struct wdm_device *desc = wdm_find_device(intf);
/*
* we notify everybody using poll of
* an exceptional situation
* must be done before recovery lest a spontaneous
* message from the device is lost
*/
spin_lock_irq(&desc->iuspin);
set_bit(WDM_RESETTING, &desc->flags); /* inform read/write */
set_bit(WDM_READ, &desc->flags); /* unblock read */
clear_bit(WDM_IN_USE, &desc->flags); /* unblock write */
desc->rerr = -EINTR;
spin_unlock_irq(&desc->iuspin);
wake_up_all(&desc->wait);
mutex_lock(&desc->rlock);
mutex_lock(&desc->wlock);
kill_urbs(desc);
cancel_work_sync(&desc->rxwork);
cancel_work_sync(&desc->service_outs_intr);
return 0;
}
static int wdm_post_reset(struct usb_interface *intf)
{
struct wdm_device *desc = wdm_find_device(intf);
int rv;
clear_bit(WDM_OVERFLOW, &desc->flags);
clear_bit(WDM_RESETTING, &desc->flags);
rv = recover_from_urb_loss(desc);
mutex_unlock(&desc->wlock);
mutex_unlock(&desc->rlock);
return rv;
}
static struct usb_driver wdm_driver = {
.name = "cdc_wdm",
.probe = wdm_probe,
.disconnect = wdm_disconnect,
#ifdef CONFIG_PM
.suspend = wdm_suspend,
.resume = wdm_resume,
.reset_resume = wdm_resume,
#endif
.pre_reset = wdm_pre_reset,
.post_reset = wdm_post_reset,
.id_table = wdm_ids,
.supports_autosuspend = 1,
USB: Disable hub-initiated LPM for comms devices. Hub-initiated LPM is not good for USB communications devices. Comms devices should be able to tell when their link can go into a lower power state, because they know when an incoming transmission is finished. Ideally, these devices would slam their links into a lower power state, using the device-initiated LPM, after finishing the last packet of their data transfer. If we enable the idle timeouts for the parent hubs to enable hub-initiated LPM, we will get a lot of useless LPM packets on the bus as the devices reject LPM transitions when they're in the middle of receiving data. Worse, some devices might blindly accept the hub-initiated LPM and power down their radios while they're in the middle of receiving a transmission. The Intel Windows folks are disabling hub-initiated LPM for all USB communications devices under a xHCI USB 3.0 host. In order to keep the Linux behavior as close as possible to Windows, we need to do the same in Linux. Set the disable_hub_initiated_lpm flag for for all USB communications drivers. I know there aren't currently any USB 3.0 devices that implement these class specifications, but we should be ready if they do. Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Cc: Marcel Holtmann <marcel@holtmann.org> Cc: Gustavo Padovan <gustavo@padovan.org> Cc: Johan Hedberg <johan.hedberg@gmail.com> Cc: Hansjoerg Lipp <hjlipp@web.de> Cc: Tilman Schmidt <tilman@imap.cc> Cc: Karsten Keil <isdn@linux-pingi.de> Cc: Peter Korsgaard <jacmet@sunsite.dk> Cc: Jan Dumon <j.dumon@option.com> Cc: Petko Manolov <petkan@users.sourceforge.net> Cc: Steve Glendinning <steve.glendinning@smsc.com> Cc: "John W. Linville" <linville@tuxdriver.com> Cc: Kalle Valo <kvalo@qca.qualcomm.com> Cc: "Luis R. Rodriguez" <mcgrof@qca.qualcomm.com> Cc: Jouni Malinen <jouni@qca.qualcomm.com> Cc: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Cc: Senthil Balasubramanian <senthilb@qca.qualcomm.com> Cc: Christian Lamparter <chunkeey@googlemail.com> Cc: Brett Rudley <brudley@broadcom.com> Cc: Roland Vossen <rvossen@broadcom.com> Cc: Arend van Spriel <arend@broadcom.com> Cc: "Franky (Zhenhui) Lin" <frankyl@broadcom.com> Cc: Kan Yan <kanyan@broadcom.com> Cc: Dan Williams <dcbw@redhat.com> Cc: Jussi Kivilinna <jussi.kivilinna@mbnet.fi> Cc: Ivo van Doorn <IvDoorn@gmail.com> Cc: Gertjan van Wingerde <gwingerde@gmail.com> Cc: Helmut Schaa <helmut.schaa@googlemail.com> Cc: Herton Ronaldo Krzesinski <herton@canonical.com> Cc: Hin-Tak Leung <htl10@users.sourceforge.net> Cc: Larry Finger <Larry.Finger@lwfinger.net> Cc: Chaoming Li <chaoming_li@realsil.com.cn> Cc: Daniel Drake <dsd@gentoo.org> Cc: Ulrich Kunitz <kune@deine-taler.de> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
2012-04-23 11:08:51 -06:00
.disable_hub_initiated_lpm = 1,
};
module_usb_driver(wdm_driver);
MODULE_AUTHOR(DRIVER_AUTHOR);
MODULE_DESCRIPTION(DRIVER_DESC);
MODULE_LICENSE("GPL");