;4c> 8b
02 4c 39 c7 75 52 4c 8b 40 08 4c 39 c7 75 66 48 89 50 08
RIP [] __list_del_entry+0x29/0xc0
RSP
CR2:
---[ end trace 99fc090ab3ff6cbc ]---
Fixes: f1bddbb ("usb: gadget: Fix binding to UDC via configfs
interface")
Signed-off-by: Felix Hädicke
Tested-by: Krzyszto
Hi,
> On 12/01/2016 10:44 PM, Felix Hädicke wrote:
> > Hi,
> >
> >> Hi,
> >>
> >> Good catch but I have a few comments to commit message:
> >>
> >> On 12/01/2016 12:24 AM, Felix Hädicke wrote:
> >>> This fixes a regre
Hi,
> Hi!
>
> I've got the following error report while running the syzkaller fuzzer.
>
> On commit 3c49de52d5647cda8b42c4255cf8a29d1e22eff5 (Dec 2).
>
> general protection fault: [#1] SMP KASAN
> Dumping ftrace buffer:
>(ftrace buffer empty)
> Modules linked in:
> CPU: 0 PID: 10564 Comm:
Hi,
> Hi,
>
> Good catch but I have a few comments to commit message:
>
> On 12/01/2016 12:24 AM, Felix Hädicke wrote:
>> This fixes a regression which was introduced by commit f1bddbb, by
>> reverting a small fragment of commit 855ed04.
>>
> Not exactly. T
Hello,
> I had expected that writing "dummy_udc.0" to
> /sys/kernel/config/usb_gadget/gser/UDC would fail, because the UDC is
> already bound to another gadget configuration
> (/sys/kernel/config/usb_gadget/acm/UDC). However, this doesn't give me
> an error, but the ACM configuration remains activ
;4c> 8b
02 4c 39 c7 75 52 4c 8b 40 08 4c 39 c7 75 66 48 89 50 08
RIP [] __list_del_entry+0x29/0xc0
RSP
CR2: 0000
---[ end trace 99fc090ab3ff6cbc ]---
Signed-off-by: Felix Hädicke
---
drivers/usb/gadget/udc/core.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff -
Hello,
in 4.8.11 and 4.9-rc7, I get an Oops (see below), when I try to activate
and afterward deactivate a second gadget configfs configuration, like this:
mkdir /sys/kernel/config/usb_gadget/acm
cd /sys/kernel/config/usb_gadget/acm
echo 0x0200 > bcdUSB
echo 0x0525 > idVendor
echo 0xa4a7 > idProd
Hello,
in 4.8.11 and 4.9-rc7, I get an Oops (see below), when I try to activate
and afterward deactivate a second gadget configfs configuration, like this:
mkdir /sys/kernel/config/usb_gadget/acm
cd /sys/kernel/config/usb_gadget/acm
echo 0x0200 > bcdUSB
echo 0x0525 > idVendor
echo 0xa4a7 > idProd
; This also fixes the following static checker warning:
> drivers/usb/gadget/function/f_fs.c:3152 ffs_func_req_match()
> warn: always true condition '(((creq->wIndex)) >= 0) => (0-u16max >= 0)'
>
> Reported-by: Dan Carpenter
> Signed-off-by: Felix Hädicke
> ---
---[ end trace 5b3336a407e1698c ]---
Signed-off-by: Felix Hädicke
Tested-by: Peter Chen
---
Changes in v2:
- Add Tested-by: Peter Chen
- Add oops dump
drivers/usb/gadget/function/u_serial.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/usb/gadget/function
ok, thanks for the hint. I dind't know this.
Just ignore this patch.
Am 16.11.2016 um 22:42 schrieb Sergei Shtylyov:
> Hello.
>
> On 11/16/2016 11:46 PM, Felix Hädicke wrote:
>
>> There can be cases when members of the gscons_info struct are used
>> uninitialized, e
There can be cases when members of the gscons_info struct are used
uninitialized, e. g. in, gserial_console_exit(), if gs_console_setup()
was not called before.
Signed-off-by: Felix Hädicke
---
drivers/usb/gadget/function/u_serial.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
Add check for NULL before calling kthread_stop().
There were cases in which gserial_console_exit() was called, but the
console thread was not started. This resulted in an invalid
kthread_stop(NULL) call.
Signed-off-by: Felix Hädicke
---
drivers/usb/gadget/function/u_serial.c | 3 ++-
1 file
Hi,
For me, it is always reproducible when CONFIG_U_SERIAL_CONSOLE is enabled.
But I think that GDB is misleading, it has probably nothing to do with
gs_buf_free(). For me, it looks like the real problem is that in
gserial_console_exit(), kthread_stop() is called on the (in my case)
uninitialised
Hello,
In 4.8.6 and 4.9-rc5, the gadget serial driver crashes during
deinitialisation when compiled with CONFIG_U_SERIAL_CONSOLE.
Steps to reproduce:
modprobe g-serial
rmmod g-serial
The problem also occurs when using configfs, when the UDC is unbound.
I does not make a difference if I use my P
warning:
drivers/usb/gadget/function/f_fs.c:3152 ffs_func_req_match()
warn: always true condition '(((creq->wIndex)) >= 0) => (0-u16max >= 0)'
Reported-by: Dan Carpenter
Signed-off-by: Felix Hädicke
---
drivers/usb/gadget/function/f_fs.c | 8
1 file changed, 4 inse
Hello Dan Carpenter,
thanks for this. It is indeed a parentheses error. I am currently
testing a patch to correct this.
Regards,
Felix
> Hello Felix Hädicke,
>
> The patch 54dfce6d07b0: "usb: gadget: f_fs: handle control requests
> not directed to interface or endpoint" fr
> This is actually not for a f_accessory replacement. The control requests are
> sent when the device is in any other gadget mode, for sending metadata and to
> trigger switching to Open Accessory Gadget Mode.
>
> The Accessory mode itself can be handled very easily with f_serial (or a very
> s
Hello,
> > I am developing a driver, which I would like to implement using
> > FunctionFS, for using the Android Open Accessory protocol on
> > non-Android devices. These accessories send some non-standard control
> > requests which are described here:
> > https://source.android.com/devices/access
Hello,
> Hi,
>
> Felix Hädicke writes:
> > It can sometimes be necessary for gadget drivers to process non-standard
> > control requests, which host devices can send without having sent
> > USB_REQ_SET_CONFIGURATION.
>
> got an example?
There are Android O
Hello,
> Hi,
>
> Felix Hädicke writes:
> > Introduces a new FunctionFS descriptor flag named
> > FUNCTIONFS_ALL_CTRL_RECIP. When this flag is enabled, control requests,
> > which are not explicitly directed to an interface or endpoint, can be
> > handle
the first usb_function instance whose req_match() returns
true.
Signed-off-by: Felix Hädicke
---
drivers/usb/gadget/composite.c | 16 ++--
drivers/usb/gadget/function/f_fs.c | 9 +++--
drivers/usb/gadget/function/f_printer.c | 6 +-
include/linux/usb/c
Introduces a new FunctionFS descriptor flag named
FUNCTIONFS_CONFIG0_SETUP.
When this flag is enabled, FunctionFS userspace drivers can process
non-standard control requests in configuration 0.
Signed-off-by: Felix Hädicke
---
drivers/usb/gadget/function/f_fs.c | 5 +++--
include/uapi/linux
-by: Felix Hädicke
---
drivers/usb/gadget/function/f_fs.c | 34 ++
include/uapi/linux/usb/functionfs.h | 1 +
2 files changed, 31 insertions(+), 4 deletions(-)
diff --git a/drivers/usb/gadget/function/f_fs.c
b/drivers/usb/gadget/function/f_fs.c
index 73515d5
24 matches
Mail list logo
