scsi_transport_srp: Reduce failover time
authorBart Van Assche <bart.vanassche@sandisk.com>
Mon, 18 May 2015 11:24:38 +0000 (13:24 +0200)
committerDoug Ledford <dledford@redhat.com>
Mon, 18 May 2015 17:35:55 +0000 (13:35 -0400)
commite68ca75200fed4700471fb84ea0d09e25f269654
tree9e3edf242fac69f72abe78a9228629bce7da1cc8
parenta44074f14ba1ea0747ea737026eb929b81993dc3
scsi_transport_srp: Reduce failover time

Unlike FC, there is no risk that SCSI devices will be offlined
by the SCSI error handler if the SCSI error handler is started
while a reconnect attempt is ongoing. Hence report timeout errors
as soon as possible if a higher software layer (e.g. multipathd)
needs to be informed about a timeout. It is assumed that if both
fast_io_fail_tmo < 0 and rport->dev_loss_tmo < 0 that this means
that there is no need to report timeouts quickly.

Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: James Bottomley <JBottomley@Odin.com>
Cc: Sagi Grimberg <sagig@mellanox.com>
Cc: Sebastian Parschauer <sebastian.riemer@profitbricks.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
drivers/scsi/scsi_transport_srp.c