On Fri, Nov 23, 2018 at 04:26:38PM +0800, Lijun Ou wrote: > This patch updates the implementation of the mailbox command > interface by using command queue instead of operating registers. > With this update, the software can be well decoupled with the > hardware. > > Signed-off-by: Lijun Ou <oulijun@xxxxxxxxxx> > Signed-off-by: Yixian Liu <liuyixian@xxxxxxxxxx> > Signed-off-by: Wei Hu (Xavier) <xavier.huwei@xxxxxxxxxx> > --- > drivers/infiniband/hw/hns/hns_roce_common.h | 3 -- > drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 69 ++++++++++++++++++----------- > drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 28 +++++++----- > 3 files changed, 59 insertions(+), 41 deletions(-) This doesn't apply: Applying: RDMA/hns: Init qp context when modify qp from reset to init Applying: RDMA/hns: Fix the bug while use multi-hop of pbl Applying: RDMA/hns: Update posting & querying mailbox error: sha1 information is lacking or useless (drivers/infiniband/hw/hns/hns_roce_hw_v2.c). error: could not build fake ancestor Patch failed at 0003 RDMA/hns: Update posting & querying mailbox All series have to be based against for-next or explicitly say what their dependencies are. If they depend on other unaccepted series then you should wait to send them unless there is good reason. It is helpful if you publish a git tree that has every series you are submitting as a branch, this will allow my 'git am' to resolve trivial context fuzz automatically, instead of hard failing. Jason
