linux/drivers/infiniband
Jack Wang d024f27de1 RDMA/ipoib: Distribute cq completion vector better
Currently ipoib choose cq completion vector based on port number, when HCA
only have one port, all the interface recv queue completion are bind to cq
completion vector 0.

To better distribute the load, use same method as __ib_alloc_cq_any to
choose completion vector, with the change, each interface now use
different completion vectors.

Link: https://lore.kernel.org/r/20201013074342.15867-1-jinpu.wang@cloud.ionos.com
Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Reviewed-by: Gioh Kim <gi-oh.kim@cloud.ionos.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2020-11-20 16:18:59 -04:00
..
core RDMA/core: remove use of dma_virt_ops 2020-11-17 15:22:07 -04:00
hw Merge branch 'for-rc' into rdma.git 2020-11-17 15:20:26 -04:00
sw RDMA/core: remove use of dma_virt_ops 2020-11-17 15:22:07 -04:00
ulp RDMA/ipoib: Distribute cq completion vector better 2020-11-20 16:18:59 -04:00
Kconfig RMDA/sw: Don't allow drivers using dma_virt_ops on highmem configs 2020-11-12 13:27:41 -04:00
Makefile treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00