> >
> > You are right, but the problem is the request->status is wrong. If the data
> > send out but report caller as -EINTR, it will introduce duplicate-send
> > issue.
> >
>
> Why -EINTR, the kernel-doc said it should return -ECONNRESET for active
> request, see include/linux/usb/gadget.h.
>
>
On Tue, Jan 05, 2016 at 04:09:47AM +, Du, Changbin wrote:
> > > To avoid this, just dequeue the request first. After usb_ep_dequeue, the
> > > request must be done or canceled.
> > >
> > > With this change, we can ensure no race condition in f_fs driver. But
> > > actually I found some of the u
Hi,
My name is Jeffrey Skoll, a philanthropist and the founder of one of the
largest private foundations in the world. I believe strongly in ‘giving while
living.’ I had one idea that never changed in my mind — that you should use
your wealth to help people and I have decided to secretly give US
> > To avoid this, just dequeue the request first. After usb_ep_dequeue, the
> > request must be done or canceled.
> >
> > With this change, we can ensure no race condition in f_fs driver. But
> > actually I found some of the udc driver has analogical issue in its
> > dequeue implementation. For ex
On Tue, Dec 29, 2015 at 02:36:58PM +0800, changbin...@intel.com wrote:
> From: "Du, Changbin"
>
> ffs_epfile_io and ffs_epfile_io_complete runs in different context, but
> there is no synchronization between them.
>
> consider the following scenario:
> 1) ffs_epfile_io interrupted by sigal while
From: Hayes Wang
Date: Mon, 4 Jan 2016 14:38:46 +0800
> When the reset_resume() is called, the flag of SELECTIVE_SUSPEND should be
> cleared and reinitialize the device, whether the SELECTIVE_SUSPEND is set
> or not. If reset_resume() is called, it means the power supply is cut or the
> device is
On Mon, 2016-01-04 at 19:07 -0300, Geyslan G. Bem wrote:
> 2016-01-04 18:52 GMT-03:00 Sergei Shtylyov :
> > > > > > This patch fixes coding style issues reported by checkpatch
> > > > > > concerning
> > > > > > to unnecessary space after a cast.
> > > > > This is a case where checkpatch is wrong,
On Mon, Nov 02, 2015 at 02:50:40AM +0100, Rafael J. Wysocki wrote:
>
> I've queued up this series for the second half of the v4.4 merge window.
>
> Thanks,
> Rafael
>
>
> ___
> linux-arm-kernel mailing list
> linux-arm-ker...@lists.infradead.org
> htt
Dear Oliver,
> You cannot honor GFP_NOIO
GFP_NOIO is used in original kernel space transmission relating sk_buf.
It is not used for userspace transmission.
I added 'if' statement around the block to skip the block where GFP_NOIO
is used for kernel space transmission.
> the cleanest solution woul
This patch reduces number of event handling threads to one.
In existing implementation, event kernel threads are created for each
port. The functions of the threads are terminationg connection and
error handling. It's too expensive to have to each port.
With this patch, event handler is a singl
This patch extends number of ports limitation in application (vhci)
side.
To do so, vhci driver supports multiple host controllers. The number of
controllers can be specified as a module parameter 'num_controllers'.
The default is 1.
ex) # insmod vhci_hcd.ko num_controllers=4
Also, ports per
This series of patches extends number of ports limitaion in application
(vhci) side.
0. Version info
v2)
# Added static to some functions and variables not called from other
files.
1. Overview
This series conatins 2 patches.
1/2:
Extends number of ports using multiple host controllers.
To introduce some application protocols like WebSocket, this patch
allows to substitute connection establishment and termination. In
combination with previous patch, both connection and transmission can
be replaced.
usbip_connection_operations_t includes open and close operation. Open
method r
This patch allows to substitute send, receive and shutdown routines for
both a) request/response PDUs among utilities and b) user space URBs
transmission.
usbip_sock_t is introduced instead of sockfd. it includes function
pointers of send/receive/shutdown routines, an argument for the
routines
Addition to README regarding user space URBs transmission.
Signed-off-by: Nobuo Iwata
---
tools/usb/usbip/README | 22 ++
1 file changed, 22 insertions(+)
diff --git a/tools/usb/usbip/README b/tools/usb/usbip/README
index 74f4afb..6b61da5 100644
--- a/tools/usb/usbip/README
Originally, USB/IP transmits requests and response PDUs for preparation
to transfer URBs in user space, after completion of the preparation,
URBs are transmitted in kernel space.
To make easy to introduce application network protocols like WebSocket,
the driver, usbip_ux.ko, forwards URBs to US
To utilize core parts of USB/IP to application protocol
implementations, this patch derives libraries by exposing some
functions of utilities and removing some unnecessary portions.
Following functions are exposed.
For command:
- usbip_attach_device()
- usbip_detach_port()
- usbip_bind_device()
This patch adds utilities transmit packets via WebSocket protocol.
WebSocket version of utilities as following.
usbws : command
usbwsa : application-side daemon
usbwsd : device-side daemon
The command supports all sub-command (ie. list, connect, disconnect,
port, bind, unbind, attach and de
This patch adds 'const' qualifier to 'char*' arguments of library
interfaces to make acceptable std::string.c_str(). Essentially, these
qualifiers are better to be used even if not to use C++. Although, I
just added to functions related to previous patch.
Also, it changes C++ reserved words (ie
USB/IP supports a function to import USB devices from application-side
machine by attach command.
The usage is as following.
dev:# (Physically attach your USB device.)
dev:# insmod usbip-core.ko and usbip-host.ko
dev:# usbipd -D
// Start usbip daemon.
dev:# usbip list -l
// List local USB devices
Originally, USB/IP transmits requests and response PDUs for preparation
to transfer URBs in user space, after the preparation, URBs are
transmitted in kernel space.
To make easy to introduce application network protocols like WebSocket
and so on, the driver, usbip_ux.ko, forwards URBs to USB/IP
This patch adds function and usage of export to README and manuals.
The wording, 'server' and 'client' is changed also.
For existing attach command, the daemon runs device side machine and
attach command is executed in application side machine. Then 'server'
is used for device side and 'client'
stub_shutdown_connection() : drivers/usb/usbip/stub_dev.c
stub_device_cleanup_urbs() : drivers/usb/usbip/stub_main.c
requests to kill pending URBs and clears priv lists.
stub_complete() : drivers/usb/usbip/stub_tx.c might be called with URBs
to have been requested to kill.
To avoid kernel
Dear all,
This series of patches introduces WebSocket to USB/IP.
0. Version info
v6)
# Added __rcu annotation to a RCU pointer to clear sparse warnings.
# Corrected a copy to RCU pointer with rcu_rcu_assign_pointer().
# Added __user annotations to arguments of read/write method.
# Added stati
2016-01-04 18:01 GMT-03:00 Alan Stern :
> On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
>
>> This patch fixes a coding style issue reported by checkpatch related to
>> many leading tabs, removing a 'do while' loop and making use of goto tag
>> instead.
>
> This is highly questionable. It's a big amou
2016-01-04 18:00 GMT-03:00 Alan Stern :
> On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
>
>> Functions must have the opening brace at the beginning of the next line
>> and body conforming indentation.
>
> This isn't necessary if the function is an empty static inline void
> routine.
Ok. It's more relat
2016-01-04 18:52 GMT-03:00 Sergei Shtylyov :
> Hello.
>
> On 01/05/2016 12:49 AM, Greg Kroah-Hartman wrote:
>
> This patch fixes coding style issues reported by checkpatch concerning
> to unnecessary space after a cast.
This is a case where checkpatch is wrong, IMO. Casts sh
On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
> 2016-01-04 17:50 GMT-03:00 Alan Stern :
> > On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
> >
> >> This patch fixes an unsigned comparison to less than 0.
> >
> > No, it doesn't. It changes an unsigned comparison for less than or
> > equal to 0, which is ve
Hello.
On 01/05/2016 12:49 AM, Greg Kroah-Hartman wrote:
This patch fixes coding style issues reported by checkpatch concerning
to unnecessary space after a cast.
This is a case where checkpatch is wrong, IMO. Casts should always be
followed by a space. I will not accept this patch.
Ok. I
On Mon, Jan 04, 2016 at 06:40:38PM -0300, Geyslan G. Bem wrote:
> 2016-01-04 17:58 GMT-03:00 Alan Stern :
> > On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
> >
> >> This patch fixes coding style issues reported by checkpatch concerning
> >> to unnecessary space after a cast.
> >
> > This is a case wher
In ffs_epfile_io error label points to a return path which includes
a kfree(data) call. However, at the beginning of the function data is
always NULL so some of the early ‘goto error’ can safely be replaced
with a trivial return statement.
Signed-off-by: Michal Nazarewicz
---
drivers/usb/gadget
From: "Du, Changbin"
ffs_epfile_io and ffs_epfile_io_complete runs in different context, but
there is no synchronization between them.
consider the following scenario:
1) ffs_epfile_io interrupted by sigal while
wait_for_completion_interruptible
2) then ffs_epfile_io set ret to -EINTR
3) just be
In the ffs_epfile_io function, data buffer is allocated for non-halt
requests. Later, after grabing a mutex, the function checks that
epfile->ep is still ep and if it’s not, it set ret to -ESHUTDOWN and
follow a path including spin_unlock_irq (just after ‘ret = -ESHUTDOWN’),
mutex_unlock (after if
In the AIO path, if allocating of a request failse, the function simply
goes to the error_lock path whose end result is returning value of ret.
However, at this point ret’s value is zero (assigned as return value from
ffs_mutex_lock).
Fix by adding ‘ret = -ENOMEM’ statement.
Signed-off-by: Michal
Eliminate one of the return paths by using a ‘goto error_mutex’ and
rearrange some if-bodies which results in reduction of the indention level
and thus hopefully makes the function easier to read and reason about.
Signed-off-by: Michal Nazarewicz
---
drivers/usb/gadget/function/f_fs.c | 127
2016-01-04 17:58 GMT-03:00 Alan Stern :
> On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
>
>> This patch fixes coding style issues reported by checkpatch concerning
>> to unnecessary space after a cast.
>
> This is a case where checkpatch is wrong, IMO. Casts should always be
> followed by a space. I
2016-01-04 17:50 GMT-03:00 Alan Stern :
> On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
>
>> This patch fixes an unsigned comparison to less than 0.
>
> No, it doesn't. It changes an unsigned comparison for less than or
> equal to 0, which is very different from less than 0.
You're right. The stateman
On 4 January 2016 at 21:39, Philipp Zabel wrote:
> Am Samstag, den 19.12.2015, 11:55 +0100 schrieb Hans de Goede:
>> On 18-12-15 12:08, Maxime Ripard wrote:
>> > - If the reset line is in a !exclusive use with more than 1 user,
>> >the refcount is modified and an error is returned to
On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
> This patch fixes a coding style issue reported by checkpatch related to
> many leading tabs, removing a 'do while' loop and making use of goto tag
> instead.
This is highly questionable. It's a big amount of code churn, nearly
impossible to verify vi
On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
> Functions must have the opening brace at the beginning of the next line
> and body conforming indentation.
This isn't necessary if the function is an empty static inline void
routine.
Alan Stern
> This patch also reduces qh_lines() header definition
On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
> This patch fixes coding style issues reported by checkpatch concerning
> to unnecessary space after a cast.
This is a case where checkpatch is wrong, IMO. Casts should always be
followed by a space. I will not accept this patch.
This must be somethi
On Mon, 4 Jan 2016, Geyslan G. Bem wrote:
> This patch fixes an unsigned comparison to less than 0.
No, it doesn't. It changes an unsigned comparison for less than or
equal to 0, which is very different from less than 0.
> Signed-off-by: Geyslan G. Bem
> ---
>
> Notes:
> I'm not sure abo
Am Samstag, den 19.12.2015, 11:55 +0100 schrieb Hans de Goede:
> On 18-12-15 12:08, Maxime Ripard wrote:
[...]
> > I guess we could also have something like this:
> >
> >* The driver gets the reference to the reset line using
> > reset_control_get or its shared variant.
> >
> > - If y
On Tue, Dec 29 2015, changbin...@intel.com wrote:
> From: "Du, Changbin"
>
> ffs_epfile_io and ffs_epfile_io_complete runs in different context, but
> there is no synchronization between them.
>
> consider the following scenario:
> 1) ffs_epfile_io interrupted by sigal while
> wait_for_completion_
This patch fixes coding style issues reported by checkpatch concerning
to unnecessary space after a cast.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.
This patch fixes coding style issues reported by checkpatch.
Coding style demands usage of C89-style comments and a specific format
when it's multiline.
This also removes the Free Software Foundation address because FSF can
change it again.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/eh
This patch fixes coding style issues reported by checkpatch. The only
change in this patch that isn't just removing spaces before opening
square brackets is at line 213 where the initialization of fls_strings[]
is placed in same line.
Signed-off-by: Geyslan G. Bem
---
Notes:
Before this patc
This patch fixes coding style issues reported by checkpatch. The vast
majority of changes in this patch are removing spaces before opening
parenthesis, but in some cases, a few additional changes are made to fix
other coding style issues.
These additional changes are:
- Spaces around >> on line
This patch fixes an unsigned comparison to less than 0.
Signed-off-by: Geyslan G. Bem
---
Notes:
I'm not sure about that comparison because in qh_lines() temp receives
the snprintf() return and thereafter occurs this comparison:
if (size < temp)
temp = size;
This patch fixes coding style issues reported by checkpatch concerning
to missing spaces around operators.
There is an additional change on line 49 that removes unnecessary
parenthesis around ternary operands.
Signed-off-by: Geyslan G. Bem
---
Notes:
Tested by compilation only.
drivers/us
This patch puts the closing parenthesis at the statement end removing
unnecessary "new line".
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 12
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
i
This patch converts macros into inline functions since the usage of
second is encouraged by Coding Style instead of the first.
Macros converted to functions:
- dbg_status
- dbg_cmd
- dbg_port
- speed_char
The size after changes remains the same.
Before:
text data bss dec hex filename
369
This patch fixes a coding style issue reported by checkpatch related to
kmalloc_array usage.
On same line the sizeof operand was enclosed by parenthesis.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb
This patch fixes coding style issues reported by checkpatch concerning
to missing line after struct declarations.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index a
This patch fixes coding style issue reported by checkpatch concerning to
an unnecessary line before close brace.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index a42b05
This patch fixes coding style issues reported by checkpatch concerning
to missing line after variable declarations.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index db
This patch fixes a coding style issue reported by checkpatch concerning
to usage of sizeof operand as a variable instead the type.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/ehci-dbg.c b/drive
This patch fixes a coding style issue reported by checkpatch related to
many leading tabs, removing a 'do while' loop and making use of goto tag
instead.
Others changes in this patch are:
- Some multiline statements are reduced (718, 729, 780, 786, 790).
- A constant is moved to right on line 7
This patch fixes coding style issues reported by checkpatch concerning
to conditional blocks without braces.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/e
Functions must have the opening brace at the beginning of the next line
and body conforming indentation.
This patch also reduces qh_lines() header definition to two lines.
Signed-off-by: Geyslan G. Bem
---
drivers/usb/host/ehci-dbg.c | 44 +---
1 file cha
This patch fixes coding style issues reported by checkpatch concerning
to switch case statements. There are few additional changes made to fix
other coding styles issues.
These additional changes are:
- The compound statement "({...})" on line 474 is pulled out from
snprintf parameters.
- O
Hi,
Bin Liu writes:
> Hi,
>
> On Mon, Jan 4, 2016 at 12:22 PM, Felipe Balbi wrote:
>>
>> Hi,
>>
>> Bin Liu writes:
>>> Hi,
>>>
>>> On Fri, Dec 18, 2015 at 5:02 AM, Uwe Kleine-König
>>> wrote:
The phy's init routine must be called before it can be used. Do so in
musb_init_controller
On nominal execution, private data allocated on port_probe and attach
are never freed. Add port_remove and release callbacks to free them
respectively.
Signed-off-by: Mathieu OTHACEHE
---
Changes in v2:
* Move release callback after attach callback
drivers/usb/serial/mxu11x0.c | 20 +++
Hi,
Here are the follow up commits proposed during last Johan review of the
new mxu11x0 driver. I also patched a memory leak on usb_serial private data.
In v2, I fixed the trivial issues reported by Johan in commits 1/4 and 4/4.
About commit 2/4, I was wrong in last cover-letter message.
The exac
The function mxu1_set_termios is called twice in open callback.
Only the first call is necessary.
Signed-off-by: Mathieu OTHACEHE
---
drivers/usb/serial/mxu11x0.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/usb/serial/mxu11x0.c b/drivers/usb/serial/mxu11x0.c
index 7ad2727..6326
Sending OPEN and START commands twice is not necessary for this driver.
Also send STOP command at close.
Signed-off-by: Mathieu OTHACEHE
---
Changes in v2:
* Only remove OPEN_PORT and START_PORT in open callback.
drivers/usb/serial/mxu11x0.c | 21 +++--
1 file changed, 7 insert
Move interrupt in endpoint test and firmware download to a new probe
callback. This avoids unnecessary memory allocations done by core
before port_probe callback is called.
If the device has to be reseted (firmware downloaded) or if the
interface is incorrect (no interrupt in endpoint), the probe
On Mon, 2016-01-04 at 12:32 +0700, Lars Melin wrote:
> On 2016-01-04 05:00, Jose Alonso wrote:
> >
> > The program Heimdall (http://glassechidna.com.au/heimdall) is used
> > to flash a new firmware in Samsung Mobile Phones. It uses only the
> > library libusb to access the device.
> >
> > The mod
Hi,
On Mon, Jan 4, 2016 at 12:22 PM, Felipe Balbi wrote:
>
> Hi,
>
> Bin Liu writes:
>> Hi,
>>
>> On Fri, Dec 18, 2015 at 5:02 AM, Uwe Kleine-König
>> wrote:
>>> The phy's init routine must be called before it can be used. Do so in
>>> musb_init_controller and the matching shutdown in musb_remo
Hi,
Bin Liu writes:
> Hi,
>
> On Fri, Dec 18, 2015 at 5:02 AM, Uwe Kleine-König
> wrote:
>> The phy's init routine must be called before it can be used. Do so in
>> musb_init_controller and the matching shutdown in musb_remove.
>>
>> Signed-off-by: Uwe Kleine-König
>> ---
>> Note I'm not entir
Hi,
On Fri, Dec 18, 2015 at 5:02 AM, Uwe Kleine-König
wrote:
> The phy's init routine must be called before it can be used. Do so in
> musb_init_controller and the matching shutdown in musb_remove.
>
> Signed-off-by: Uwe Kleine-König
> ---
> Note I'm not entirely sure this is the right place, bu
On Mon, 2016-01-04 at 18:23 +0100, Oliver Neukum wrote:
> On Mon, 2016-01-04 at 23:06 +0700, Lars Melin wrote:
> > On 2016-01-04 13:43, Oliver Neukum wrote:
> > > On Mon, 2016-01-04 at 12:32 +0700, Lars Melin wrote:
> > K
> > > >
> > > > 04e8:6601 is also used for Samsung SCH-U209 CDMA modem so ca
On Mon, 2016-01-04 at 23:06 +0700, Lars Melin wrote:
> On 2016-01-04 13:43, Oliver Neukum wrote:
> > On Mon, 2016-01-04 at 12:32 +0700, Lars Melin wrote:
> K
> >>
> >> 04e8:6601 is also used for Samsung SCH-U209 CDMA modem so can not be
> >> ignored.
> >> 04e8:685d is an update mode id so it can be
On 2016-01-04 13:43, Oliver Neukum wrote:
On Mon, 2016-01-04 at 12:32 +0700, Lars Melin wrote:
K
04e8:6601 is also used for Samsung SCH-U209 CDMA modem so can not be
ignored.
04e8:685d is an update mode id so it can be ignored by cdc-acm.
04e8:68c3 does not look like update mode, it has cdc-ac
On Mon, 4 Jan 2016, Jia-Ju Bai wrote:
> The driver calls pci_set_mwi to enable memory-write-invalidate when it
> is initialized, but does not call pci_clear_mwi when it is removed. Many
> other drivers calls pci_clear_mwi when pci_set_mwi is called, such as
> r8169, 8139cp and e1000.
>
> This
On Mon, 4 Jan 2016, Jia-Ju Bai wrote:
> The driver calls ehci_mem_init to allocate memory resources.
> But these resources are not freed when ehci_halt fails.
>
> This patch adds "ehci_mem_cleanup" in error handling code to fix this problem.
>
> Signed-off-by: Jia-Ju Bai
> ---
> drivers/usb/h
On Wed, Dec 30, 2015 at 09:15:55AM +, Pathak, Rahul (R.) wrote:
> Hello Greg,
>
> This patch is not yet merged, its already been reviewed and acked by Alan.
>
Relax. Everyone is on vacation. This stuff is not life or death that
it must be done right away.
regards,
dan carpenter
--
To uns
On Sun, Jan 03, 2016 at 06:11:07PM -0500, Edwin Olson wrote:
> Hi Johan et al,
>
> Now that I understand the behavior, and know how to work-around the
> issue, I won't press too hard for functionality changes. But perhaps
> the documentation can be improved by suggesting that the custom
> divisor
On Mon, 2016-01-04 at 12:06 +0100, Johan Hovold wrote:
> Now, we could extend this long-since established behaviour, but I'm
> not
> sure it's worth it given that any further calls using the hung up tty
> device will also return with errno set to -EIO by the tty layer. And
> this we can't easily ch
On Mon, Jan 04, 2016 at 09:09:55AM +0100, Oliver Neukum wrote:
> On Sun, 2016-01-03 at 17:22 +0100, Johan Hovold wrote:
> > On Mon, Dec 21, 2015 at 02:17:22PM +0100, Oliver Neukum wrote:
> > > Disconnecting a device is not just a hang up. The device is gone.
> > > We should tell user space the trut
On Sun, 2016-01-03 at 17:22 +0100, Johan Hovold wrote:
> On Mon, Dec 21, 2015 at 02:17:22PM +0100, Oliver Neukum wrote:
> > Disconnecting a device is not just a hang up. The device is gone.
> > We should tell user space the truth immediately.
>
> I'm not sure about this one. The usb-serial drivers
The driver calls pci_set_mwi to enable memory-write-invalidate when it
is initialized, but does not call pci_clear_mwi when it is removed. Many
other drivers calls pci_clear_mwi when pci_set_mwi is called, such as
r8169, 8139cp and e1000.
This patch adds a function "ehci_pci_remove" to remove t
The driver calls ehci_mem_init to allocate memory resources.
But these resources are not freed when ehci_halt fails.
This patch adds "ehci_mem_cleanup" in error handling code to fix this problem.
Signed-off-by: Jia-Ju Bai
---
drivers/usb/host/ehci-hcd.c |4 +++-
1 file changed, 3 insertion
83 matches
Mail list logo
