On Wed, Feb 12, 2020 at 09:26:26AM +0200, Leon Romanovsky wrote: > From: Leon Romanovsky <leonro@xxxxxxxxxxxx> > > Hi, > > This pack of small fixes is sent as a patchset simply to simplify their > tracking. Some of them, like first and second patches were already > sent to the mailing list. The ucma patch was in our regression for whole > cycle and we didn't notice any failures related to that change. > > Changelog of second patch: > 1. Maor added IB_QP_PKEY_INDEX and IB_QP_PORT checks and I rewrote the > code logic to be less hairy. > > Thanks > > Leon Romanovsky (2): > RDMA/ucma: Mask QPN to be 24 bits according to IBTA I put this one in to for-next > RDMA/mlx5: Prevent overflow in mmap offset calculations > > Maor Gottlieb (1): > RDMA/core: Fix protection fault in get_pkey_idx_qp_list > > Michael Guralnik (1): > RDMA/core: Add missing list deletion on freeing event queue > > Yishai Hadas (1): > IB/mlx5: Fix async events cleanup flows > > Yonatan Cohen (1): > IB/umad: Fix kernel crash while unloading ib_umad > > Zhu Yanjun (1): > RDMA/rxe: Fix soft lockup problem due to using tasklets in softirq And these to for-rc > Parav Pandit (1): > Revert "RDMA/cma: Simplify rdma_resolve_addr() error flow" > > Valentine Fatiev (1): > IB/ipoib: Fix double free of skb in case of multicast traffic in CM > mode These I want to see the discussion on Thanks, Jason
