3.16.40-rc1 review patch. If anyone has any objections, please let me know.
------------------
From: Moshe Lazer <moshel@xxxxxxxxxxxx>
commit 6bc1a656ab9f57f0112823b4a36930c9a29d1f89 upstream.
When calling reg_mr of large MRs (e.g. 4GB) from multiple processes
and MR caches can't supply the required amount of MRs the slow-path
of MR allocation may be used. In this case we need to serialize the
slow-path between the processes to avoid soft lock.
Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Moshe Lazer <moshel@xxxxxxxxxxxx>
Signed-off-by: Maor Gottlieb <maorg@xxxxxxxxxxxx>
Reviewed-by: Eli Cohen <eli@xxxxxxxxxxxx>
Signed-off-by: Leon Romanovsky <leon@xxxxxxxxxx>
Signed-off-by: Doug Ledford <dledford@xxxxxxxxxx>
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
drivers/infiniband/hw/mlx5/mlx5_ib.h | 2 ++
drivers/infiniband/hw/mlx5/mr.c | 6 +++++-
2 files changed, 7 insertions(+), 1 deletion(-)
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -376,6 +376,8 @@ struct mlx5_ib_dev {
struct mlx5_ib_resources devr;
struct mlx5_mr_cache cache;
struct timer_list delay_timer;
+ /* Prevents soft lock on massive reg MRs */
+ struct mutex slow_path_mutex;
int fill_delay;
};
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -554,6 +554,7 @@ int mlx5_mr_cache_init(struct mlx5_ib_de
int err;
int i;
+ mutex_init(&dev->slow_path_mutex);
cache->wq = create_singlethread_workqueue("mkey_cache");
if (!cache->wq) {
mlx5_ib_warn(dev, "failed to create work queue\n");
@@ -909,9 +910,12 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct
}
}
- if (!mr)
+ if (!mr) {
+ mutex_lock(&dev->slow_path_mutex);
mr = reg_create(pd, virt_addr, length, umem, ncont, page_shift,
access_flags);
+ mutex_unlock(&dev->slow_path_mutex);
+ }
if (IS_ERR(mr)) {
err = PTR_ERR(mr);