Re: [PATCH] rwsem: add rwsem_is_contended

[Peter Hurley <peter@hurleysoftware.com>]

On 09/01/2013 04:32 AM, Michel Lespinasse wrote:
> Hi Josef,
>
> On Fri, Aug 30, 2013 at 7:14 AM, Josef Bacik <jbacik@fusionio.com> wrote:
>> Btrfs uses an rwsem to control access to its extent tree.  Threads will hold a
>> read lock on this rwsem while they scan the extent tree, and if need_resched()
>> they will drop the lock and schedule.  The transaction commit needs to take a
>> write lock for this rwsem for a very short period to switch out the commit
>> roots.  If there are a lot of threads doing this caching operation we can starve
>> out the committers which slows everybody out.  To address this we want to add
>> this functionality to see if our rwsem has anybody waiting to take a write lock
>> so we can drop it and schedule for a bit to allow the commit to continue.
>> Thanks,
>>
>> Signed-off-by: Josef Bacik <jbacik@fusionio.com>
>
> FYI, I once tried to introduce something like this before, but my use
> case was pretty weak so it was not accepted at the time. I don't think
> there were any objections to the API itself though, and I think it's
> potentially a good idea if you use case justifies it.

Exactly, I'm concerned about the use case: readers can't starve writers.
Of course, lots of existing readers can temporarily prevent a writer from
acquiring, but those readers would already have the lock. Any new readers
wouldn't be able to prevent a waiting writer from obtaining the lock.

Josef,
Could you be more explicit, maybe with some detailed numbers about the
condition you report?

I say that because a subtle bug that could mistakenly wait a reader
existed in the rwsem implementation until relatively recently. Is there
some other lurking problem?

> Two comments:
>
> - Note that there are two rwsem implementations - if you are going to
> add functionality to rwsem.h you want to add the same functionality in
> rwsem-spinlock.h as well.
>
> - I would prefer if you could avoid taking the wait_lock in your
> rwsem.h implementation. In your use case (read lock is known to be
> held), checking for sem->count < 0 would be sufficient to indicate a
> writer is queued (or getting onto the queue). In the general case,
> some architectures have the various values set up so that
> RWSEM_WAITING_BIAS != RWSEM_ACTIVE_WRITE_BIAS - for these
> architectures at least, you can check for waiters by looking if the
> lowest bit of RWSEM_WAITING_BIAS is set in sem->count.

Michel,

I'm glad you point out a much better approach --- but why are we
considering open-coding down_read_trylock()/down_write_trylock?

Regards,
Peter Hurley