RDMA/mlx5: Delay emptying a cache entry when a new MR is added to it recently

[ Upstream commit d6793ca97b76642b77629dd0783ec64782a50bdb ]

Fixing a typo that causes a cache entry to shrink immediately after adding
to it new MRs if the entry size exceeds the high limit.  In doing so, the
cache misses its purpose to prevent the creation of new mkeys on the
runtime by using the cached ones.

Fixes: b9358bdbc713 ("RDMA/mlx5: Fix locking in MR cache work queue")
Link: https://lore.kernel.org/r/fcb546986be346684a016f5ca23a0567399145fa.1627370131.git.leonro@nvidia.com
Signed-off-by: Aharon Landau <aharonl@nvidia.com>
Reviewed-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 971694e781b653cb5346f4a2c55d045526860535..19346693c1da4a3b6068b6aeb3de2feda3227e89 100644 (file)
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -526,8 +526,8 @@ static void __cache_work_func(struct mlx5_cache_ent *ent)
                 */
                spin_unlock_irq(&ent->lock);
                need_delay = need_resched() || someone_adding(cache) ||
-                            time_after(jiffies,
-                                       READ_ONCE(cache->last_add) + 300 * HZ);
+                            !time_after(jiffies,
+                                        READ_ONCE(cache->last_add) + 300 * HZ);
                spin_lock_irq(&ent->lock);
                if (ent->disabled)
                        goto out;