On 10/18/21 3:40 PM, Max Gurtovoy wrote:

Error recovery mechanism is duplicated in RDMA and TCP transports. Move
this logic to common code.

Also update the RDMA/TCP transport drivers to use this API and remove
the duplicated code.

Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Israel Rukshin <redacted>
Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
---
  drivers/nvme/host/fabrics.c | 10 ++++++++++
  drivers/nvme/host/fabrics.h |  1 +
  drivers/nvme/host/rdma.c    | 25 ++++++++-----------------
  drivers/nvme/host/tcp.c     | 19 +++++--------------
  4 files changed, 24 insertions(+), 31 deletions(-)

_Technically_ we shouldn't start reconnecting for command errors; 
reconnection only should start once KATO expires.

But that might be a topic for a different discussion.

Reviewed-by: Hannes Reinecke <hare@suse.de>

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                Kernel Storage Architect
hare@suse.de                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Felix Imendörffer