Hi Jason, Sorry, I didn't notice that the first three patches of this series had been applied to you for-next branch, so I sent this new version. Need I resend the last two patches in a new series? Thanks Weihang On 2020/3/10 19:22, Weihang Li wrote: > Process of wqe is hard to understand and maintain in hns, this series > refactor wqe related code, especially in hns_roce_v2_post_send(). > > Previous discussion could be found at: > - v2: https://patchwork.kernel.org/cover/11422963/ > - v1: https://patchwork.kernel.org/cover/11415395/ > > Changes since v2: > - Remove a redundant assignment of "NULL" in patch 4/5. > > Changes since v1: > - Fix comments from Leon about the inplementation of to_hr_opcode() in > patch 3/5. > - Patch 4/5 couldn't be applied. Just do a rebase. > > Xi Wang (5): > RDMA/hns: Rename wqe buffer related functions > RDMA/hns: Optimize wqe buffer filling process for post send > RDMA/hns: Optimize the wr opcode conversion from ib to hns > RDMA/hns: Optimize base address table config flow for qp buffer > RDMA/hns: Optimize wqe buffer set flow for post send > > drivers/infiniband/hw/hns/hns_roce_device.h | 10 +- > drivers/infiniband/hw/hns/hns_roce_hem.c | 16 +- > drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 9 +- > drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 588 ++++++++++++++-------------- > drivers/infiniband/hw/hns/hns_roce_qp.c | 48 +-- > 5 files changed, 319 insertions(+), 352 deletions(-) >
