mbox series

[v2,for-next,00/10] RDMA/hns: Various cleanups

Message ID 1584674622-52773-1-git-send-email-liweihang@huawei.com (mailing list archive)
Headers show
Series RDMA/hns: Various cleanups | expand

Message

Weihang Li March 20, 2020, 3:23 a.m. UTC 
This series contains some cleanups for the hns driver:
- patch 1 unifies format of prints in hns_roce_hw_v2.c and hns_roce_pd.c.
- patch 2 ~ 5 are some simple modifications.
- patch 6 ~ 10 remove some dead codes.

Previous discussions can be found at:
https://patchwork.kernel.org/cover/11447213/

Changes since v1:
- Drop patch 3/11 from series v1 because it's wrong to use IS_ERR_OR_NULL
  on a pointer returned by kmalloc().

Lang Cheng (4):
  RDMA/hns: Simplify attribute judgment code
  RDMA/hns: Adjust the qp status value sequence of the hardware
  RDMA/hns: Remove definition of cq doorbell structure
  RDMA/hns: Remove redundant qpc setup operations

Lijun Ou (2):
  RDMA/hns: Unify format of prints
  RDMA/hns: Optimize hns_roce_alloc_vf_resource()

Weihang Li (3):
  RDMA/hns: Fix a wrong judgment of return value
  RDMA/hns: Remove redundant assignment of wc->smac when polling cq
  RDMA/hns: Remove redundant judgment of qp_type

Wenpeng Liang (1):
  RDMA/hns: Remove meaningless prints

 drivers/infiniband/hw/hns/hns_roce_hw_v1.c |   9 +-
 drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 560 ++++++++---------------------
 drivers/infiniband/hw/hns/hns_roce_hw_v2.h |   7 +-
 drivers/infiniband/hw/hns/hns_roce_mr.c    |   2 +-
 drivers/infiniband/hw/hns/hns_roce_pd.c    |   6 +-
 5 files changed, 159 insertions(+), 425 deletions(-)

Comments

Jason Gunthorpe March 26, 2020, 7:55 p.m. UTC | #1 
On Fri, Mar 20, 2020 at 11:23:32AM +0800, Weihang Li wrote:
> This series contains some cleanups for the hns driver:
> - patch 1 unifies format of prints in hns_roce_hw_v2.c and hns_roce_pd.c.
> - patch 2 ~ 5 are some simple modifications.
> - patch 6 ~ 10 remove some dead codes.
> 
> Previous discussions can be found at:
> https://patchwork.kernel.org/cover/11447213/
> 
> Changes since v1:
> - Drop patch 3/11 from series v1 because it's wrong to use IS_ERR_OR_NULL
>   on a pointer returned by kmalloc().
> 
> Lang Cheng (4):
>   RDMA/hns: Simplify attribute judgment code
>   RDMA/hns: Adjust the qp status value sequence of the hardware
>   RDMA/hns: Remove definition of cq doorbell structure
>   RDMA/hns: Remove redundant qpc setup operations
> 
> Lijun Ou (2):
>   RDMA/hns: Unify format of prints
>   RDMA/hns: Optimize hns_roce_alloc_vf_resource()
> 
> Weihang Li (3):
>   RDMA/hns: Fix a wrong judgment of return value
>   RDMA/hns: Remove redundant assignment of wc->smac when polling cq
>   RDMA/hns: Remove redundant judgment of qp_type
> 
> Wenpeng Liang (1):
>   RDMA/hns: Remove meaningless prints

Applied to for-next, thanks

Jason