OSDN Git Service

RDMA/bnxt_re: Fix the page_size used during the MR creation
authorSelvin Xavier <selvin.xavier@broadcom.com>
Sun, 7 May 2023 18:29:29 +0000 (11:29 -0700)
committerJason Gunthorpe <jgg@nvidia.com>
Fri, 12 May 2023 21:30:30 +0000 (18:30 -0300)
commit08c7f09356e45d093d1867c7a3c6ac6526e2f98b
tree4ac85a45a53835bfdb55f838968a2c69f3b5319e
parentac9a78681b921877518763ba0e89202254349d1b
RDMA/bnxt_re: Fix the page_size used during the MR creation

Driver populates the list of pages used for Memory region wrongly when
page size is more than system page size. This is causing a failure when
some of the applications that creates MR with page size as 2M.  Since HW
can support multiple page sizes, pass the correct page size while creating
the MR.

Also, driver need not adjust the number of pages when HW Queues are
created with user memory. It should work with the number of dma blocks
returned by ib_umem_num_dma_blocks. Fix this calculation also.

Fixes: 0c4dcd602817 ("RDMA/bnxt_re: Refactor hardware queue memory allocation")
Fixes: f6919d56388c ("RDMA/bnxt_re: Code refactor while populating user MRs")
Link: https://lore.kernel.org/r/1683484169-9539-1-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
drivers/infiniband/hw/bnxt_re/qplib_res.c
drivers/infiniband/hw/bnxt_re/qplib_sp.c