From: Michal Nazarewicz <mina86@mina86.com>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: Ivaylo Dimitrov <ivo.g.dimitrov.75@gmail.com>,
        Tony Lindgren <tony@atomide.com>, linux-kernel@vger.kernel.org,
        linux-usb@vger.kernel.org, Felipe Balbi <felipe.balbi@linux.intel.com>,
        Bin Liu <b-liu@ti.com>, pali.rohar@gmail.com
Subject: Re: [PATCH] usb: f_mass_storage: test whether thread is running before starting another
In-Reply-To: <Pine.LNX.4.44L0.1604051340030.2233-100000@iolanthe.rowland.org>
Organization: http://mina86.com/
References: <Pine.LNX.4.44L0.1604051340030.2233-100000@iolanthe.rowland.org>
User-Agent: Notmuch/0.19+53~g2e63a09 (http://notmuchmail.org) Emacs/25.1.50.1 (x86_64-unknown-linux-gnu)
Face: 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
Date: Tue, 05 Apr 2016 20:07:45 +0200
Message-ID: <xa1ty48svsi6.fsf@mina86.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8BIT
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1754
Lines: 38

> On Tue, 5 Apr 2016, Michal Nazarewicz wrote:
>> When binding the function to usb_configuration, check whether the thread
>> is running before starting another one.  Without that, when function
>> instance is added to multiple configurations, fsg_bing starts multiple
>> threads with all but the latest one being forgotten by the driver.  This
>> leads to obvious thread leaks, possible lockups when trying to halt the
>> machine and possible more issues.
>> 
>> This fixes issues with legacy/multi¹ gadget as well as configfs gadgets
>> when mass_storage function is added to multiple configurations.
>> 
>> This change also simplifies API since the legacy gadgets no longer need
>> to worry about starting the thread by themselves (which was where bug
>> in legacy/multi was in the first place).
>> 
>> ¹ I have no example failure though.  Conclusion that legacy/multi has
>>   a bug is based purely on me reading the code.
>> 
>> Signed-off-by: Michal Nazarewicz <mina86@mina86.com>

On Tue, Apr 05 2016, Alan Stern wrote:
> This doesn't address the problem I raised in a previous email.  
> Sharing one thread among several function instances in the same config
> will not work if one of them encounters an error.

Each usb_function_instance has its own fsg_common and its own thread.
This was true in the past and is true with this patch as well.

And unless I’m missing something, sharing a thread among multiple
usb_function’s does not prevent the driver from working correctly.

Having the thread run even when it’s not used may be considered wasteful
but that’s an orthogonal issue to the configfs failure.

-- 
Best regards
ミハウ “????????????????86” ナザレヴイツ
«If at first you don’t succeed, give up skydiving»