Messages in this thread | ![/](/images/icornerl.gif) | | From | Jason Gunthorpe <> | Date | Wed, 27 Jul 2022 15:17:55 -0300 | Subject | Re: [PATCH rdma-next v1 0/5] MR cache cleanup |
| |
On Tue, Jul 26, 2022 at 10:19:06AM +0300, Michael Guralnik wrote: > Hi, > > In this series, Aharon continues to clean mlx5 MR cache logic. > > Thanks > > v1: Change push_mkey to eliminate locking on mail flow > v0: http://lore.kernel.org/all/cover.1654601897.git.leonro@nvidia.com/ > > Aharon Landau (5): > RDMA/mlx5: Replace ent->lock with xa_lock > RDMA/mlx5: Replace cache list with Xarray > RDMA/mlx5: Store the number of in_use cache mkeys instead of total_mrs > RDMA/mlx5: Store in the cache mkeys instead of mrs > RDMA/mlx5: Rename the mkey cache variables and functions
Applied to for-next, thanks
Jason
| ![\](/images/icornerr.gif) |