* [PATCH] ksmbd: call ib_drain_qp when disconnected
@ 2022-10-02 3:01 Namjae Jeon
2022-10-02 18:11 ` Tom Talpey
0 siblings, 1 reply; 4+ messages in thread
From: Namjae Jeon @ 2022-10-02 3:01 UTC (permalink / raw)
To: linux-cifs
Cc: smfrench, senozhatsky, tom, atteh.mailbox, Namjae Jeon, Hyunchul Lee
When disconnected, call ib_drain_qp to cancel all pending work requests
and prevent ksmbd_conn_handler_loop from waiting for a long time
for those work requests to compelete.
Signed-off-by: Hyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
---
fs/ksmbd/transport_rdma.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/fs/ksmbd/transport_rdma.c b/fs/ksmbd/transport_rdma.c
index 0315bca3d53b..096eda9ef873 100644
--- a/fs/ksmbd/transport_rdma.c
+++ b/fs/ksmbd/transport_rdma.c
@@ -1527,6 +1527,8 @@ static int smb_direct_cm_handler(struct rdma_cm_id *cm_id,
}
case RDMA_CM_EVENT_DEVICE_REMOVAL:
case RDMA_CM_EVENT_DISCONNECTED: {
+ ib_drain_qp(t->qp);
+
t->status = SMB_DIRECT_CS_DISCONNECTED;
wake_up_interruptible(&t->wait_status);
wake_up_interruptible(&t->wait_reassembly_queue);
--
2.25.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] ksmbd: call ib_drain_qp when disconnected
2022-10-02 3:01 [PATCH] ksmbd: call ib_drain_qp when disconnected Namjae Jeon
@ 2022-10-02 18:11 ` Tom Talpey
2022-10-03 14:31 ` Namjae Jeon
0 siblings, 1 reply; 4+ messages in thread
From: Tom Talpey @ 2022-10-02 18:11 UTC (permalink / raw)
To: Namjae Jeon, linux-cifs
Cc: smfrench, senozhatsky, atteh.mailbox, Hyunchul Lee
On 10/1/2022 11:01 PM, Namjae Jeon wrote:
> When disconnected, call ib_drain_qp to cancel all pending work requests
> and prevent ksmbd_conn_handler_loop from waiting for a long time
> for those work requests to compelete.
>
> Signed-off-by: Hyunchul Lee <hyc.lee@gmail.com>
> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
> ---
> fs/ksmbd/transport_rdma.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/fs/ksmbd/transport_rdma.c b/fs/ksmbd/transport_rdma.c
> index 0315bca3d53b..096eda9ef873 100644
> --- a/fs/ksmbd/transport_rdma.c
> +++ b/fs/ksmbd/transport_rdma.c
> @@ -1527,6 +1527,8 @@ static int smb_direct_cm_handler(struct rdma_cm_id *cm_id,
> }
> case RDMA_CM_EVENT_DEVICE_REMOVAL:
> case RDMA_CM_EVENT_DISCONNECTED: {
> + ib_drain_qp(t->qp);
> +
> t->status = SMB_DIRECT_CS_DISCONNECTED;
> wake_up_interruptible(&t->wait_status);
> wake_up_interruptible(&t->wait_reassembly_queue);
Because we're now flushing the cancelled work requests, don't
we need to also wake up &t->wait_send_pending, to abort any
waiters on the send wq?
Tom.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] ksmbd: call ib_drain_qp when disconnected
2022-10-02 18:11 ` Tom Talpey
@ 2022-10-03 14:31 ` Namjae Jeon
2022-10-03 15:48 ` Tom Talpey
0 siblings, 1 reply; 4+ messages in thread
From: Namjae Jeon @ 2022-10-03 14:31 UTC (permalink / raw)
To: Tom Talpey; +Cc: linux-cifs, smfrench, senozhatsky, atteh.mailbox, Hyunchul Lee
2022-10-03 3:11 GMT+09:00, Tom Talpey <tom@talpey.com>:
> On 10/1/2022 11:01 PM, Namjae Jeon wrote:
>> When disconnected, call ib_drain_qp to cancel all pending work requests
>> and prevent ksmbd_conn_handler_loop from waiting for a long time
>> for those work requests to compelete.
>>
>> Signed-off-by: Hyunchul Lee <hyc.lee@gmail.com>
>> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
>> ---
>> fs/ksmbd/transport_rdma.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/fs/ksmbd/transport_rdma.c b/fs/ksmbd/transport_rdma.c
>> index 0315bca3d53b..096eda9ef873 100644
>> --- a/fs/ksmbd/transport_rdma.c
>> +++ b/fs/ksmbd/transport_rdma.c
>> @@ -1527,6 +1527,8 @@ static int smb_direct_cm_handler(struct rdma_cm_id
>> *cm_id,
>> }
>> case RDMA_CM_EVENT_DEVICE_REMOVAL:
>> case RDMA_CM_EVENT_DISCONNECTED: {
>> + ib_drain_qp(t->qp);
>> +
>> t->status = SMB_DIRECT_CS_DISCONNECTED;
>> wake_up_interruptible(&t->wait_status);
>> wake_up_interruptible(&t->wait_reassembly_queue);
>
> Because we're now flushing the cancelled work requests, don't
> we need to also wake up &t->wait_send_pending, to abort any
> waiters on the send wq?
Could you please elaborate more how it could be a problem ?
send_pending is decreased and wake_up &t->wait_send_pending if it is
zero in send_done().
>
> Tom.
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] ksmbd: call ib_drain_qp when disconnected
2022-10-03 14:31 ` Namjae Jeon
@ 2022-10-03 15:48 ` Tom Talpey
0 siblings, 0 replies; 4+ messages in thread
From: Tom Talpey @ 2022-10-03 15:48 UTC (permalink / raw)
To: Namjae Jeon
Cc: linux-cifs, smfrench, senozhatsky, atteh.mailbox, Hyunchul Lee
On 10/3/2022 10:31 AM, Namjae Jeon wrote:
> 2022-10-03 3:11 GMT+09:00, Tom Talpey <tom@talpey.com>:
>> On 10/1/2022 11:01 PM, Namjae Jeon wrote:
>>> When disconnected, call ib_drain_qp to cancel all pending work requests
>>> and prevent ksmbd_conn_handler_loop from waiting for a long time
>>> for those work requests to compelete.
>>>
>>> Signed-off-by: Hyunchul Lee <hyc.lee@gmail.com>
>>> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
>>> ---
>>> fs/ksmbd/transport_rdma.c | 2 ++
>>> 1 file changed, 2 insertions(+)
>>>
>>> diff --git a/fs/ksmbd/transport_rdma.c b/fs/ksmbd/transport_rdma.c
>>> index 0315bca3d53b..096eda9ef873 100644
>>> --- a/fs/ksmbd/transport_rdma.c
>>> +++ b/fs/ksmbd/transport_rdma.c
>>> @@ -1527,6 +1527,8 @@ static int smb_direct_cm_handler(struct rdma_cm_id
>>> *cm_id,
>>> }
>>> case RDMA_CM_EVENT_DEVICE_REMOVAL:
>>> case RDMA_CM_EVENT_DISCONNECTED: {
>>> + ib_drain_qp(t->qp);
>>> +
>>> t->status = SMB_DIRECT_CS_DISCONNECTED;
>>> wake_up_interruptible(&t->wait_status);
>>> wake_up_interruptible(&t->wait_reassembly_queue);
>>
>> Because we're now flushing the cancelled work requests, don't
>> we need to also wake up &t->wait_send_pending, to abort any
>> waiters on the send wq?
> Could you please elaborate more how it could be a problem ?
> send_pending is decreased and wake_up &t->wait_send_pending if it is
> zero in send_done().
I took another look at __ib_drain_sq() and yes it does put the
qp into ERR, and only steals the single WR that it itself posted.
As long as we're not using IB_POLL_DIRECT (which ksmbd isn't), we're
good.
Reviewed-by: Tom Talpey <tom@talpey.com>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2022-10-03 15:48 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-02 3:01 [PATCH] ksmbd: call ib_drain_qp when disconnected Namjae Jeon
2022-10-02 18:11 ` Tom Talpey
2022-10-03 14:31 ` Namjae Jeon
2022-10-03 15:48 ` Tom Talpey
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.