* [PATCH] nvme-rdma: Don't update queue count when failing to set io queues
@ 2021-07-28 9:41 Ruozhu Li
2021-07-29 8:30 ` liruozhu
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Ruozhu Li @ 2021-07-28 9:41 UTC (permalink / raw)
To: linux-nvme; +Cc: sagi, lengchao
We update ctrl->queue_count and schedule another reconnect when io queue
count is zero.But we will never try to create any io queue in next reco-
nnection, because ctrl->queue_count already set to zero.We will end up
having an admin-only session in Live state, which is exactly what we try
to avoid in the original patch.
Update ctrl->queue_count after queue_count zero checking to fix it.
Signed-off-by: Ruozhu Li <liruozhu@huawei.com>
---
drivers/nvme/host/rdma.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 7f6b3a991501..3bd9cbc80246 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -735,13 +735,13 @@ static int nvme_rdma_alloc_io_queues(struct nvme_rdma_ctrl *ctrl)
if (ret)
return ret;
- ctrl->ctrl.queue_count = nr_io_queues + 1;
- if (ctrl->ctrl.queue_count < 2) {
+ if (nr_io_queues == 0) {
dev_err(ctrl->ctrl.device,
"unable to set any I/O queues\n");
return -ENOMEM;
}
+ ctrl->ctrl.queue_count = nr_io_queues + 1;
dev_info(ctrl->ctrl.device,
"creating %d I/O queues.\n", nr_io_queues);
--
2.16.4
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] nvme-rdma: Don't update queue count when failing to set io queues
2021-07-28 9:41 [PATCH] nvme-rdma: Don't update queue count when failing to set io queues Ruozhu Li
@ 2021-07-29 8:30 ` liruozhu
2021-08-06 20:11 ` Sagi Grimberg
2021-08-10 16:24 ` Christoph Hellwig
2 siblings, 0 replies; 5+ messages in thread
From: liruozhu @ 2021-07-29 8:30 UTC (permalink / raw)
To: linux-nvme
I realize that there is a problem with this modification, which may
cause ctrl->queue_count to be inconsistent with the actual number when
the connection is deleted.
I will resend a patch and try to explain the problem more clearly.
在 2021/7/28 17:41, Ruozhu Li 写道:
> We update ctrl->queue_count and schedule another reconnect when io queue
> count is zero.But we will never try to create any io queue in next reco-
> nnection, because ctrl->queue_count already set to zero.We will end up
> having an admin-only session in Live state, which is exactly what we try
> to avoid in the original patch.
> Update ctrl->queue_count after queue_count zero checking to fix it.
>
> Signed-off-by: Ruozhu Li <liruozhu@huawei.com>
> ---
> drivers/nvme/host/rdma.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
> index 7f6b3a991501..3bd9cbc80246 100644
> --- a/drivers/nvme/host/rdma.c
> +++ b/drivers/nvme/host/rdma.c
> @@ -735,13 +735,13 @@ static int nvme_rdma_alloc_io_queues(struct nvme_rdma_ctrl *ctrl)
> if (ret)
> return ret;
>
> - ctrl->ctrl.queue_count = nr_io_queues + 1;
> - if (ctrl->ctrl.queue_count < 2) {
> + if (nr_io_queues == 0) {
> dev_err(ctrl->ctrl.device,
> "unable to set any I/O queues\n");
> return -ENOMEM;
> }
>
> + ctrl->ctrl.queue_count = nr_io_queues + 1;
> dev_info(ctrl->ctrl.device,
> "creating %d I/O queues.\n", nr_io_queues);
>
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] nvme-rdma: Don't update queue count when failing to set io queues
2021-07-28 9:41 [PATCH] nvme-rdma: Don't update queue count when failing to set io queues Ruozhu Li
2021-07-29 8:30 ` liruozhu
@ 2021-08-06 20:11 ` Sagi Grimberg
2021-08-07 3:21 ` liruozhu
2021-08-10 16:24 ` Christoph Hellwig
2 siblings, 1 reply; 5+ messages in thread
From: Sagi Grimberg @ 2021-08-06 20:11 UTC (permalink / raw)
To: Ruozhu Li, linux-nvme; +Cc: lengchao
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Can you send this patch for tcp as well?
On 7/28/21 2:41 AM, Ruozhu Li wrote:
> We update ctrl->queue_count and schedule another reconnect when io queue
> count is zero.But we will never try to create any io queue in next reco-
> nnection, because ctrl->queue_count already set to zero.We will end up
> having an admin-only session in Live state, which is exactly what we try
> to avoid in the original patch.
> Update ctrl->queue_count after queue_count zero checking to fix it.
>
> Signed-off-by: Ruozhu Li <liruozhu@huawei.com>
> ---
> drivers/nvme/host/rdma.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
> index 7f6b3a991501..3bd9cbc80246 100644
> --- a/drivers/nvme/host/rdma.c
> +++ b/drivers/nvme/host/rdma.c
> @@ -735,13 +735,13 @@ static int nvme_rdma_alloc_io_queues(struct nvme_rdma_ctrl *ctrl)
> if (ret)
> return ret;
>
> - ctrl->ctrl.queue_count = nr_io_queues + 1;
> - if (ctrl->ctrl.queue_count < 2) {
> + if (nr_io_queues == 0) {
> dev_err(ctrl->ctrl.device,
> "unable to set any I/O queues\n");
> return -ENOMEM;
> }
>
> + ctrl->ctrl.queue_count = nr_io_queues + 1;
> dev_info(ctrl->ctrl.device,
> "creating %d I/O queues.\n", nr_io_queues);
>
>
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] nvme-rdma: Don't update queue count when failing to set io queues
2021-08-06 20:11 ` Sagi Grimberg
@ 2021-08-07 3:21 ` liruozhu
0 siblings, 0 replies; 5+ messages in thread
From: liruozhu @ 2021-08-07 3:21 UTC (permalink / raw)
To: Sagi Grimberg, linux-nvme; +Cc: lengchao
Thanks, I will send the tcp version of this patch right away.
On 2021/8/7 4:11, Sagi Grimberg wrote:
> Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
>
> Can you send this patch for tcp as well?
>
> On 7/28/21 2:41 AM, Ruozhu Li wrote:
>> We update ctrl->queue_count and schedule another reconnect when io queue
>> count is zero.But we will never try to create any io queue in next reco-
>> nnection, because ctrl->queue_count already set to zero.We will end up
>> having an admin-only session in Live state, which is exactly what we try
>> to avoid in the original patch.
>> Update ctrl->queue_count after queue_count zero checking to fix it.
>>
>> Signed-off-by: Ruozhu Li <liruozhu@huawei.com>
>> ---
>> drivers/nvme/host/rdma.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
>> index 7f6b3a991501..3bd9cbc80246 100644
>> --- a/drivers/nvme/host/rdma.c
>> +++ b/drivers/nvme/host/rdma.c
>> @@ -735,13 +735,13 @@ static int nvme_rdma_alloc_io_queues(struct
>> nvme_rdma_ctrl *ctrl)
>> if (ret)
>> return ret;
>> - ctrl->ctrl.queue_count = nr_io_queues + 1;
>> - if (ctrl->ctrl.queue_count < 2) {
>> + if (nr_io_queues == 0) {
>> dev_err(ctrl->ctrl.device,
>> "unable to set any I/O queues\n");
>> return -ENOMEM;
>> }
>> + ctrl->ctrl.queue_count = nr_io_queues + 1;
>> dev_info(ctrl->ctrl.device,
>> "creating %d I/O queues.\n", nr_io_queues);
>>
> .
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] nvme-rdma: Don't update queue count when failing to set io queues
2021-07-28 9:41 [PATCH] nvme-rdma: Don't update queue count when failing to set io queues Ruozhu Li
2021-07-29 8:30 ` liruozhu
2021-08-06 20:11 ` Sagi Grimberg
@ 2021-08-10 16:24 ` Christoph Hellwig
2 siblings, 0 replies; 5+ messages in thread
From: Christoph Hellwig @ 2021-08-10 16:24 UTC (permalink / raw)
To: Ruozhu Li; +Cc: linux-nvme, sagi, lengchao
Thanks,
applied to nvme-5.15.
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2021-08-10 16:31 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-07-28 9:41 [PATCH] nvme-rdma: Don't update queue count when failing to set io queues Ruozhu Li
2021-07-29 8:30 ` liruozhu
2021-08-06 20:11 ` Sagi Grimberg
2021-08-07 3:21 ` liruozhu
2021-08-10 16:24 ` Christoph Hellwig
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.