On Fri, 2018-05-04 at 10:57 +0800, Lijun Ou wrote:
> This patch series fix some bugs, include automatically
> load driver for mainstream linux distribution as well
> as some optimizations for hip08 and hip06.
>
> Lijun Ou (11):
> RDMA/hns: Add rq inline flags judgement
> RDMA/hns: Bugfix for rq record db for kernel
> RDMA/hns: Load the RoCE dirver automatically
> RDMA/hns: Update convert function of endian format
> RDMA/hns: Add return operation when configured global param fail
> RDMA/hns: Not support qp transition from reset to reset for hip06
> RDMA/hns: Fix the bug with rq sge
> RDMA/hns: Set desc_dma_addr for zero when free cmq desc
> RDMA/hns: Enable inner_pa_vld filed of mpt
> RDMA/hns: Set NULL for __internal_mr
> RDMA/hns: Fix the bug with NULL pointer
>
> drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 3 ++-
> drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 29 +++++++++++++++++++++--------
> drivers/infiniband/hw/hns/hns_roce_main.c | 2 +-
> drivers/infiniband/hw/hns/hns_roce_qp.c | 10 +++++++++-
> 4 files changed, 33 insertions(+), 11 deletions(-)
>
The commit messages here leave something to be desired. If you are
suggesting changes for the rc cycle, then each commit message should
list the problem that is fixed by the change. Several of them said
things like:
This patch mainly configure value for __internal_mr of mr_free_pd.
Well, so? Why do we care if you set that? What bug happens if we
don't?
I took this series in, but please do better with your commit messages in
the future or I'll start rejecting them when they aren't clear.
--
Doug Ledford <dledford@xxxxxxxxxx>
GPG KeyID: B826A3330E572FDD
Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDDAttachment:
signature.asc
Description: This is a digitally signed message part
