Tested-by: Yi Zhang <yi.zh...@redhat.com>

This patch fixed the issue I filed[1] which use rdma_rxe for nvme-rdma testing.

[1]
http://lists.infradead.org/pipermail/linux-nvme/2020-August/018988.html

Thanks
Yi

On 9/12/20 6:00 AM, Bart Van Assche wrote:
On 2020-09-08 19:01, Bart Van Assche wrote:
The above patch didn't compile, but the patch below does and makes the hang
disappear. So feel free to add the following to the patch below:

Tested-by: Bart Van Assche <bvanass...@acm.org>

diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index c36b4d2b61e0..23ee65a9185f 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -1285,6 +1285,8 @@ static void disable_device(struct ib_device *device)
                remove_client_context(device, cid);
        }

+       ib_cq_pool_destroy(device);
+
        /* Pairs with refcount_set in enable_device */
        ib_device_put(device);
        wait_for_completion(&device->unreg_completion);
@@ -1328,6 +1330,8 @@ static int enable_device_and_get(struct ib_device *device)
                        goto out;
        }

+       ib_cq_pool_init(device);
+
        down_read(&clients_rwsem);
        xa_for_each_marked (&clients, index, client, CLIENT_REGISTERED) {
                ret = add_client_context(device, client);
@@ -1400,7 +1404,6 @@ int ib_register_device(struct ib_device *device, const 
char *name)
                goto dev_cleanup;
        }

-       ib_cq_pool_init(device);
        ret = enable_device_and_get(device);
        dev_set_uevent_suppress(&device->dev, false);
        /* Mark for userspace that device is ready */
@@ -1455,7 +1458,6 @@ static void __ib_unregister_device(struct ib_device 
*ib_dev)
                goto out;

        disable_device(ib_dev);
-       ib_cq_pool_destroy(ib_dev);

        /* Expedite removing unregistered pointers from the hash table */
        free_netdevs(ib_dev);
Hi Jason,

Please let me know how you want to proceed with this patch.

Thanks,

Bart.


Reply via email to