Re: [PATCH v4 for-next 03/12] RDMA/core: Use refcount_t instead of atomic_t on refcount of ib_mad_snoop_private

On 2021/6/9 2:04, Jason Gunthorpe wrote:

On Fri, May 28, 2021 at 05:37:34PM +0800, Weihang Li wrote:

quoted

The refcount_t API will WARN on underflow and overflow of a reference
counter, and avoid use-after-free risks.

Signed-off-by: Weihang Li <redacted>
---
 drivers/infiniband/core/mad_priv.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/core/mad_priv.h b/drivers/infiniband/core/mad_priv.h
index 4aa16b3..25e573d 100644
--- a/drivers/infiniband/core/mad_priv.h
+++ b/drivers/infiniband/core/mad_priv.h

@@ -115,7 +115,7 @@ struct ib_mad_snoop_private {
 	struct ib_mad_qp_info *qp_info;
 	int snoop_index;
 	int mad_snoop_flags;
-	atomic_t refcount;
+	refcount_t refcount;
 	struct completion comp;
 };

Since this is never used I changed this to just delete it

Jason

Sure, thanks.

Weihang