The original code is logically correct because
spdk_nvme_qpair_process_completions() may decrease
ns_ctx->current_queue_depth. But I still recommend doing
an optimization so users can easily understand the code.
Prioritize code cleanup more than one additional polling.
Change-Id: I3613b1b4e294661fd35b7e5bebdb32e1c8352270
Signed-off-by: Hailiang Wang <hailiangx.e.wang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14332
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <qun.wan@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
95fd6cb455