Hi Linus, Rather more regressions that usual, but nothing too scary. Good to know people are testing Thanks, Jason The following changes since commit c500bee1c5b2f1d59b1081ac879d73268ab0ff17: Linux 5.14-rc4 (2021-08-01 17:04:17 -0700) are available in the Git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git tags/for-linus for you to fetch changes up to 2638a32348bbb1c384dbbd515fd2b12c155f0188: RDMA/iw_cxgb4: Fix refcount underflow while destroying cqs. (2021-08-05 22:33:00 -0300) ---------------------------------------------------------------- RDMA v5.14 second rc Pull Request Several small recent regressions: - Typo causing incorrect operation of the mlx5 mkey cache expiration - Revert a CM patch that is breaking some ULPs - Typo breaking SRQ in rxe - Revert a rxe patch breaking icrc calculation - Static checker warning about unbalance locking in hns - Subtle cxgb4 regression from a recent atomic to refcount conversion ---------------------------------------------------------------- Aharon Landau (1): RDMA/mlx5: Delay emptying a cache entry when a new MR is added to it recently Bob Pearson (2): RDMA/rxe: Use the correct size of wqe when processing SRQ RDMA/rxe: Restore setting tot_len in the IPv4 header Dakshaja Uppalapati (1): RDMA/iw_cxgb4: Fix refcount underflow while destroying cqs. Mike Marciniszyn (1): RDMA/cma: Revert INIT-INIT patch Yangyang Li (1): RDMA/hns: Fix the double unlock problem of poll_sem drivers/infiniband/core/cma.c | 17 ++++++++++++++++- drivers/infiniband/hw/cxgb4/cq.c | 12 +++++++++--- drivers/infiniband/hw/cxgb4/ev.c | 6 ++---- drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 3 ++- drivers/infiniband/hw/hns/hns_roce_cmd.c | 7 +++---- drivers/infiniband/hw/hns/hns_roce_main.c | 4 +--- drivers/infiniband/hw/mlx5/mr.c | 4 ++-- drivers/infiniband/sw/rxe/rxe_net.c | 1 + drivers/infiniband/sw/rxe/rxe_resp.c | 2 +- 9 files changed, 37 insertions(+), 19 deletions(-)