> On Mar 10, 2022, at 12:09 AM, Dai Ngo <dai.ngo@xxxxxxxxxx> wrote:
>
>
> On 3/9/22 1:46 PM, Chuck Lever III wrote:
>>
>>> On Mar 4, 2022, at 7:37 PM, Dai Ngo <dai.ngo@xxxxxxxxxx> wrote:
>>>
>>> Update nfsd_breaker_owns_lease() to handle delegation conflict
>>> with courtesy clients. If conflict was caused courtesy client
>>> then discard the courtesy client by setting CLIENT_EXPIRED and
>>> return conflict resolved. Client with CLIENT_EXPIRED is expired
>>> by the laundromat.
>>>
>>> Signed-off-by: Dai Ngo <dai.ngo@xxxxxxxxxx>
>>> ---
>>> fs/nfsd/nfs4state.c | 18 ++++++++++++++++++
>>> 1 file changed, 18 insertions(+)
>>>
>>> diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
>>> index 583ac807e98d..40a357fd1a14 100644
>>> --- a/fs/nfsd/nfs4state.c
>>> +++ b/fs/nfsd/nfs4state.c
>>> @@ -4727,6 +4727,24 @@ static bool nfsd_breaker_owns_lease(struct file_lock *fl)
>>> struct svc_rqst *rqst;
>>> struct nfs4_client *clp;
>>>
>>> + clp = dl->dl_stid.sc_client;
>>> + /*
>>> + * need to sync with courtesy client trying to reconnect using
>>> + * the cl_cs_lock, nn->client_lock can not be used since this
>>> + * function is called with the fl_lck held.
>>> + */
>>> + spin_lock(&clp->cl_cs_lock);
>>> + if (test_bit(NFSD4_CLIENT_EXPIRED, &clp->cl_flags)) {
>>> + spin_unlock(&clp->cl_cs_lock);
>>> + return true;
>>> + }
>>> + if (test_bit(NFSD4_CLIENT_COURTESY, &clp->cl_flags)) {
>>> + set_bit(NFSD4_CLIENT_EXPIRED, &clp->cl_flags);
>>> + spin_unlock(&clp->cl_cs_lock);
>>> + return true;
>>> + }
>>> + spin_unlock(&clp->cl_cs_lock);
>>> +
>> Nit: Please add nfs4_check_and_expire_courtesy_client() in this patch
>> instead of in 05/11.
>
> That means nfs4_check_and_expire_courtesy_client is being called
> in 05/11 but is not defined in 05/11. Is that ok?
I thought I saw a hunk in 5/11 that converts nfsd_breaker_owns_lease()
to use nfs4_check_and_expire_courtesy_client().
If so, I prefer if you add nfs4_check_and_expire_courtesy_client() in
this patch, and call it from nfsd_breaker_owns_lease(). Then the
additional clean-up in 5/11 isn't needed.
Again, this is a nit. But since you are driving a v16 soon anyway,
I'd like the patches changed before they are merged.
> -Dai
>
>>
>>
>>> if (!i_am_nfsd())
>>> return false;
>>> rqst = kthread_data(current);
>>> --
>>> 2.9.5
>>>
>> --
>> Chuck Lever
>>
>>
>>
--
Chuck Lever
