Re: [PATCH-tip v3 02/14] locking/rwsem: Make owner available even if !CONFIG_RWSEM_SPIN_ON_OWNER

[Waiman Long]

On 04/12/2019 10:07 AM, Waiman Long wrote:
> On 04/12/2019 10:04 AM, Waiman Long wrote:
>> On 04/12/2019 03:09 AM, Ingo Molnar wrote:
>>> * Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
>>>
>>>> On Fri, Apr 12, 2019 at 09:02:09AM +0200, Ingo Molnar wrote:
>>>>> * Waiman Long <longman@xxxxxxxxxx> wrote:
>>>>>
>>>>>> On 04/11/2019 04:12 AM, Peter Zijlstra wrote:
>>>>>>> On Wed, Apr 10, 2019 at 02:42:19PM -0400, Waiman Long wrote:
>>>>>>>> The owner field in the rw_semaphore structure is used primarily for
>>>>>>>> optimistic spinning. However, identifying the rwsem owner can also be
>>>>>>>> helpful in debugging as well as tracing locking related issues when
>>>>>>>> analyzing crash dump. The owner field may also store state information
>>>>>>>> that can be important to the operation of the rwsem.
>>>>>>>>
>>>>>>>> So the owner field is now made a permanent member of the rw_semaphore
>>>>>>>> structure irrespective of CONFIG_RWSEM_SPIN_ON_OWNER.
>>>>>>> sem->owner is still initialized under CONFIG_RWSEM_SPIN_ON_OWNER.
>>>>>> Oh, you are right. I missed that part. I will fix it in the next version.
>>>>> Could you please post the next series against tip:WIP.locking/core, which 
>>>>> is already being dogfood-ed in -tip and which I'm running on my desktop? 
>>>>> I'll backmerge any fixes as needed/requested.
>>>> Urgh, please no, that's going to be hell to review :/
>>> Ok - full patches then. I'll handle the fallout.
>>>
>>> Thanks,
>>>
>>> 	Ingo
>> I will post an updated patchset later today.
>>
>> Sorry for the omission.
>>
>> Cheers,
>> Longman
>>
> BTW, the v3 patch that I posted yesterday should work fine as long as
> CONFIG_RWSEM_SPIN_ON_OWNER is defined.
>
> Cheers,
> Longman
>
Oh, I see that the WIP.locking/core is currently merged into master. I
would say rwsem part1 patchset and patch 1 of part2 are stable. So I
would suggest merging those into the the master will be good. The rests
are still under review until I get an OK from Peter. If they miss the
next merge window and have to postpone to 5.3, I am fine with that.

Thanks,
Longman