Re: gadgetfs USB2.0 Chapter 9 Tests: Test after "Addressed State" fails

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



>>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.

On Tue, Dec 24 2013, roshan.jhalani@xxxxxxxxxxxx wrote:
> Then, I would say this bug exists in both Android 4.2 and Android 4.4
> version. But we are facing issue in Android 4.4 because kernel uses
> FunctionFs instead of f_adb.c.

That may be the case.

> Do you think this should be ONLY fixed in user space or we can avoid
> all ep releases in kernel?

I think this should be fixed in user space since kernel space behaves as
it always has and as advertised[1].  Changing the behaviour would change
the ABI and risk situations that when a user space daemon crashes,
gadget's USB configuration becomes dysfunctional for the host.

Perhaps it would make sense in some cases for the FunctionFS to hang on
even when all files are closed, but I don't think the issue you've
described justifies such a change, since it's easily fixed in user space
(a quick hack is to simply dup() ep0 and never close it).

[1] See Documentation/usb/functionfs.txt:

        When all files are closed the function disables itself.

Perhaps the documentation should be made clearer here though.

-- 
Best regards,                                         _     _
.o. | Liege of Serenely Enlightened Majesty of      o' \,=./ `o
..o | Computer Science,  Michał “mina86” Nazarewicz    (o o)
ooo +--<mpn@xxxxxxxxxx>--<xmpp:mina86@xxxxxxxxxx>--ooO--(_)--Ooo--

Attachment: signature.asc
Description: PGP signature


[Index of Archives]     [Linux Media]     [Linux Input]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [Old Linux USB Devel Archive]

  Powered by Linux