…apter

When a host adapter is attached to a specific NUMA node, allocating
scsi_device and scsi_target via kzalloc() may place them on a remote
node.  All hot-path I/O accesses to these structures then cross the NUMA
interconnect, adding latency and consuming inter-node bandwidth.

Use kzalloc_node() with dev_to_node(shost->dma_dev) so allocations land
on the same node as the HBA, reducing cross-node traffic and improving
I/O performance on NUMA systems.

Signed-off-by: James Rizzo <[email protected]>
Signed-off-by: Sumit Saxena <[email protected]>

iorequest_cnt and iodone_cnt are updated on every command dispatch and
completion, often from different CPUs on high queue depth workloads.
Using adjacent atomic_t fields caused cache line contention between the
submission and completion paths.

Represent these statistics with struct percpu_counter so increments are
mostly local to each CPU, avoiding false sharing without growing
struct scsi_device further for cache-line padding.

Suggested-by: Bart Van Assche <[email protected]>
Signed-off-by: Sumit Saxena <[email protected]>