Re: [Qemu-devel] [big lock] Discussion about the convention of device's DMA each other after breaking down biglock

[Jan Kiszka]

On 2012-09-19 11:27, Jan Kiszka wrote:
> On 2012-09-19 11:23, Avi Kivity wrote:
>> On 09/19/2012 12:19 PM, liu ping fan wrote:
>>> On Wed, Sep 19, 2012 at 5:14 PM, Paolo Bonzini <address@hidden> wrote:
>>>> Il 19/09/2012 11:11, liu ping fan ha scritto:
>>>>>>> Why not? devA will drop its local lock, devX will retake the big lock
>>>>>>> recursively, devB will take its local lock.  In the end, we have biglock
>>>>>>> -> devB.
>>>>>>>
>>>>> But when adopting local lock, we assume take local lock, then biglock.
>>>>
>>>> No, because the local lock will be dropped before taking the biglock.
>>>> The order must always be coarse->fine.
>>>>
>>> But if we takes coarse firstly, then the mmio-dispatcher will still
>>> contend for the big lock against each other.
>>
>> Can you detail the sequence?
>>
>>>
>>>> I don't know if the front-end (device) lock should come before or after
>>>> the back-end (e.g. netdev) lock in the hierarchy, but that's another story.
>>>>
>>> I think fine->coarse may be the rule, since for some code path, it is
>>> not necessary to take coarse lock.
>>
>> coarse->fine doesn't mean you have to take the coarse lock.
>>
>> Valid:
>>   lock(coarse)
>>   lock(fine)
> 
> This is invalid due to prio inversion issues, independent of potential
> deadlocks.

Err, forget it - the other way around is the problem.

Sorry,
Jan

-- 
Siemens AG, Corporate Technology, CT RTC ITP SDP-DE
Corporate Competence Center Embedded Linux