在 2024/1/26 4:02, Leon Romanovsky 写道:
On Thu, Jan 25, 2024 at 09:38:24AM -0400, Jason Gunthorpe wrote:
On Thu, Jan 25, 2024 at 08:52:57PM +0800, Junxian Huang wrote:
On 2024/1/25 20:30, Leon Romanovsky wrote:
From: Or Har-Toov <ohartoov@xxxxxxxxxx>
In the dereg flow, UMEM is not a good enough indication whether an MR
is from userspace since in mlx5_ib_rereg_user_mr there are some cases
when a new MR is created and the UMEM of the old MR is set to NULL.
Currently when mlx5_ib_dereg_mr is called on the old MR, UMEM is NULL
but cache_ent can be different than NULL. So, the mkey will not be
destroyed.
Therefore checking if mkey is from user application and cacheable
should be done by checking if rb_key or cache_ent exist and all other kind of
mkeys should be destroyed.
Fixes: dd1b913fb0d0 ("RDMA/mlx5: Cache all user cacheable mkeys on dereg MR flow")
Signed-off-by: Or Har-Toov <ohartoov@xxxxxxxxxx>
Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxx>
---
drivers/infiniband/hw/mlx5/mr.c | 15 ++++++++-------
1 file changed, 8 insertions(+), 7 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 12bca6ca4760..3c241898e064 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -1857,6 +1857,11 @@ static int cache_ent_find_and_store(struct mlx5_ib_dev *dev,
return ret;
}
+static bool is_cacheable_mkey(struct mlx5_ib_mkey mkey)
I think it's better using a pointer as the parameter instead of the struct itself.
Indeed, that looks like a typo
It is suboptimal to pass struct by value, because whole struct will be copied,
Agree. With a pointer, an address is passed. With the struct, the whole
struct is copied and passed. The overhead of calling function is less
with a pointer.
Zhu Yanjun
but it is not a mistake too.
Thanks
Thanks,
Jason
