Re: [RFC PATCH 2/2] drivers/base: add managed token devres interfaces

[Shuah Khan <shuah.kh@samsung.com>]

On 04/17/2014 02:10 PM, Tejun Heo wrote:
> On Thu, Apr 17, 2014 at 02:01:32PM -0600, Shuah Khan wrote:
>> Operating on the lock should be atomic, which is what devres_update()
>> is doing. It can be simplified as follows by holding devres_lock
>> in devm_token_lock().
>>
>> spin_lock_irqsave(&dev->devres_lock, flags);
>> if (tkn_ptr->status == TOKEN_DEVRES_FREE)
>> 	tkn_ptr->status = TOKEN_DEVRES_BUSY;
>> spin_unlock_irqrestore(&dev->devres_lock, flags);
>>
>> Is this in-line with what you have in mind?
>
> How is that different from tkn_ptr->status = TOKEN_DEVRES_BUSY?
>

I see what you are saying. The code path doesn't ensure two threads
not getting the lock. I have a bug in here that my rc settings aren't
protected. You probably noticed that the RFC tag on the patch and this
isn't fully cooked yet.

I started working on driver changes that use this token and I might have
to add owner for the token as well. I hope to work these details out and
send a real patch.

thanks,
-- Shuah

-- 
Shuah Khan
Senior Linux Kernel Developer - Open Source Group
Samsung Research America(Silicon Valley)
shuah.kh@samsung.com | (970) 672-0658