IB/mlx4: Synchronize cleanup of MCGs in MCG paravirtualization
A client re-register event invokes cleanup of all MCGs. This is
required to protect against misbehaved guests leading to corruption of
join/leave database. However, since cleaning up the MCGs is a heavy
operation, it is pushed to a work queue for further processing.
Client re-register is also propagated to ULPs (e.g IPoIB).
However, since the cleanup is performed in a workqueue, the ULP could
leave and re-join groups before the cleanup occurs. In this case,
when the cleanup takes place, it prunes the (newly-joined) MCGs and
the ULP is left without actual MCGs while believing it joined them.
Fix this by setting the flushing flag before invoking the cleanup task
and clearing it after flushing is complete.
Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c
index 44ff480..25b2cdf 100644
--- a/drivers/infiniband/hw/mlx4/mcg.c
+++ b/drivers/infiniband/hw/mlx4/mcg.c
@@ -1075,10 +1075,6 @@
unsigned long end;
int count;
- if (ctx->flushing)
- return;
-
- ctx->flushing = 1;
for (i = 0; i < MAX_VFS; ++i)
clean_vf_mcast(ctx, i);
@@ -1108,9 +1104,6 @@
force_clean_group(group);
}
mutex_unlock(&ctx->mcg_table_lock);
-
- if (!destroy_wq)
- ctx->flushing = 0;
}
struct clean_work {
@@ -1124,6 +1117,7 @@
struct clean_work *cw = container_of(work, struct clean_work, work);
_mlx4_ib_mcg_port_cleanup(cw->ctx, cw->destroy_wq);
+ cw->ctx->flushing = 0;
kfree(cw);
}
@@ -1131,13 +1125,20 @@
{
struct clean_work *work;
+ if (ctx->flushing)
+ return;
+
+ ctx->flushing = 1;
+
if (destroy_wq) {
_mlx4_ib_mcg_port_cleanup(ctx, destroy_wq);
+ ctx->flushing = 0;
return;
}
work = kmalloc(sizeof *work, GFP_KERNEL);
if (!work) {
+ ctx->flushing = 0;
mcg_warn("failed allocating work for cleanup\n");
return;
}