diff mbox series

IB/hifi1: add a null check of kzalloc_node in hfi1_ipoib_txreq_init

Message ID 20230227100212.910820-1-void0red@gmail.com (mailing list archive)
State Accepted
Headers show
Series IB/hifi1: add a null check of kzalloc_node in hfi1_ipoib_txreq_init | expand

Commit Message

void0red Feb. 27, 2023, 10:02 a.m. UTC
From: Kang Chen <void0red@gmail.com>

kzalloc_node may fails, check it and do the cleanup.

Signed-off-by: Kang Chen <void0red@gmail.com>
---
 drivers/infiniband/hw/hfi1/ipoib_tx.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

Comments

Leon Romanovsky March 13, 2023, 11:52 a.m. UTC | #1
On Mon, Feb 27, 2023 at 06:02:12PM +0800, void0red wrote:
> From: Kang Chen <void0red@gmail.com>
> 
> kzalloc_node may fails, check it and do the cleanup.
> 
> Signed-off-by: Kang Chen <void0red@gmail.com>
> ---
>  drivers/infiniband/hw/hfi1/ipoib_tx.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)

Added Fixes line and applied.
Fixes: b1151b74ff68 ("IB/hfi1: Fix alloc failure with larger txqueuelen")

Thanks

> 
> diff --git a/drivers/infiniband/hw/hfi1/ipoib_tx.c b/drivers/infiniband/hw/hfi1/ipoib_tx.c
> index 5d9a7b09c..349eb4139 100644
> --- a/drivers/infiniband/hw/hfi1/ipoib_tx.c
> +++ b/drivers/infiniband/hw/hfi1/ipoib_tx.c
> @@ -737,10 +737,13 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
>  		txq->tx_ring.shift = ilog2(tx_item_size);
>  		txq->tx_ring.avail = hfi1_ipoib_ring_hwat(txq);
>  		tx_ring = &txq->tx_ring;
> -		for (j = 0; j < tx_ring_size; j++)
> +		for (j = 0; j < tx_ring_size; j++) {
>  			hfi1_txreq_from_idx(tx_ring, j)->sdma_hdr =
>  				kzalloc_node(sizeof(*tx->sdma_hdr),
>  					     GFP_KERNEL, priv->dd->node);
> +			if (!hfi1_txreq_from_idx(tx_ring, j)->sdma_hdr)
> +				goto free_txqs;
> +		}
>  
>  		netif_napi_add_tx(dev, &txq->napi, hfi1_ipoib_poll_tx_ring);
>  	}
> -- 
> 2.34.1
>
Leon Romanovsky March 13, 2023, 11:53 a.m. UTC | #2
On Mon, 27 Feb 2023 18:02:12 +0800, void0red wrote:
> kzalloc_node may fails, check it and do the cleanup.
> 
> 

Applied, thanks!

[1/1] IB/hifi1: add a null check of kzalloc_node in hfi1_ipoib_txreq_init
      https://git.kernel.org/rdma/rdma/c/c874ad879c2f29

Best regards,
diff mbox series

Patch

diff --git a/drivers/infiniband/hw/hfi1/ipoib_tx.c b/drivers/infiniband/hw/hfi1/ipoib_tx.c
index 5d9a7b09c..349eb4139 100644
--- a/drivers/infiniband/hw/hfi1/ipoib_tx.c
+++ b/drivers/infiniband/hw/hfi1/ipoib_tx.c
@@ -737,10 +737,13 @@  int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
 		txq->tx_ring.shift = ilog2(tx_item_size);
 		txq->tx_ring.avail = hfi1_ipoib_ring_hwat(txq);
 		tx_ring = &txq->tx_ring;
-		for (j = 0; j < tx_ring_size; j++)
+		for (j = 0; j < tx_ring_size; j++) {
 			hfi1_txreq_from_idx(tx_ring, j)->sdma_hdr =
 				kzalloc_node(sizeof(*tx->sdma_hdr),
 					     GFP_KERNEL, priv->dd->node);
+			if (!hfi1_txreq_from_idx(tx_ring, j)->sdma_hdr)
+				goto free_txqs;
+		}
 
 		netif_napi_add_tx(dev, &txq->napi, hfi1_ipoib_poll_tx_ring);
 	}