>>This is a bug in user space. FUNCTIONFS_DISABLE event means the host has
>>chosen a configuration with given function disabled. User-space should clean
>>up it's state, as any pending requests of any >> kind are now invalid. It
>>should not close all of the ep files.
Then, I would say this
>> This function is called whenever an ep file is released by user space (be it
>> ep0 file or the other “regular” ep files). It is complementary to the
>> ffs_data_opened function which in tandem track whether >> at least one
>> epfile is open.
>> The way it works is that if user space closes
>> Do you have a patch that does that? It would help in analysing the issue.
>> Something quick and dirty will do just to see.
Please refer patch below. I am not so good in gadgetFs, so just tried to hack
this issue as per the following:
Signed-off-by: Roshan Jhalani
---
drivers/usb/gadget/c
Hi,
I spent some time on this issue. And it looks like the only possible solution
to avoid USB disconnect from gadgetFs driver.
In my trial, I avoided ep0_release/ep_release in every disconnect with function
state update. With this CV passes without functionality impact.
I don't know what was t
Hi,
Recently, we encountered one issue in Kernel 3.10(On Android 4.4) where USB
enumeration is falling after many connect/disconnect.
The reason for this issue is, we started using usb_gstrings_attach for string
update in configuration and cdev->next_string_id is never set to 0 in
disconnect c
Hi Macro,
We have observed same issue and reason for this issue is reset_config which
triggers complete USB disconnect from F_FS.
For SET_CONFIG(Config#0) there is no need to do USB Disconnect. This seems to
be bottleneck issue for USB compliance.
I believe this issue should be addressed by Gad
