[PATCH net v2] net/smc: fix NULL sndbuf_desc in smc_cdc_tx_handler()

[PATCH net v2] net/smc: fix NULL sndbuf_desc in smc_cdc_tx_handler()

[D. Wythe]

From: "D. Wythe" <alibuda@linux.alibaba.com>

When performing a stress test on SMC-R by rmmod mlx5_ib driver
during the wrk/nginx test, we found that there is a probability
of triggering a panic while terminating all link groups.

This issue dues to the race between smc_smcr_terminate_all()
and smc_buf_create().

			smc_smcr_terminate_all

smc_buf_create
/* init */
conn->sndbuf_desc = NULL;
...

			__smc_lgr_terminate
				smc_conn_kill
					smc_close_abort
						smc_cdc_get_slot_and_msg_send

			__softirqentry_text_start
				smc_wr_tx_process_cqe
					smc_cdc_tx_handler
						READ(conn->sndbuf_desc->len);
						/* panic dues to NULL sndbuf_desc */

conn->sndbuf_desc = xxx;

This patch tries to fix the issue by always to check the sndbuf_desc
before send any cdc msg, to make sure that no null pointer is
seen during cqe processing.

Fixes: 0b29ec643613 ("net/smc: immediate termination for SMCR link groups")
Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>

v2 -> v1: change retval from EINVAL to ENOBUFS

---
 net/smc/smc_cdc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c
index 53f63bf..89105e9 100644
--- a/net/smc/smc_cdc.c
+++ b/net/smc/smc_cdc.c
@@ -114,6 +114,9 @@ int smc_cdc_msg_send(struct smc_connection *conn,
 	union smc_host_cursor cfed;
 	int rc;
 
+	if (unlikely(!READ_ONCE(conn->sndbuf_desc)))
+		return -ENOBUFS;
+
 	smc_cdc_add_pending_send(conn, pend);
 
 	conn->tx_cdc_seq++;
-- 
1.8.3.1

Re: [PATCH net v2] net/smc: fix NULL sndbuf_desc in smc_cdc_tx_handler()

[Tony Lu]

On Wed, Mar 08, 2023 at 04:17:12PM +0800, D. Wythe wrote:
> From: "D. Wythe" <alibuda@linux.alibaba.com>
> 
> When performing a stress test on SMC-R by rmmod mlx5_ib driver
> during the wrk/nginx test, we found that there is a probability
> of triggering a panic while terminating all link groups.
> 
> This issue dues to the race between smc_smcr_terminate_all()
> and smc_buf_create().
> 
> 			smc_smcr_terminate_all
> 
> smc_buf_create
> /* init */
> conn->sndbuf_desc = NULL;
> ...
> 
> 			__smc_lgr_terminate
> 				smc_conn_kill
> 					smc_close_abort
> 						smc_cdc_get_slot_and_msg_send
> 
> 			__softirqentry_text_start
> 				smc_wr_tx_process_cqe
> 					smc_cdc_tx_handler
> 						READ(conn->sndbuf_desc->len);
> 						/* panic dues to NULL sndbuf_desc */
> 
> conn->sndbuf_desc = xxx;
> 
> This patch tries to fix the issue by always to check the sndbuf_desc
> before send any cdc msg, to make sure that no null pointer is
> seen during cqe processing.
> 
> Fixes: 0b29ec643613 ("net/smc: immediate termination for SMCR link groups")
> Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>

Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>

Thanks.

> 
> v2 -> v1: change retval from EINVAL to ENOBUFS

Better to put the changelog before Fixes tag.

> 
> ---
>  net/smc/smc_cdc.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c
> index 53f63bf..89105e9 100644
> --- a/net/smc/smc_cdc.c
> +++ b/net/smc/smc_cdc.c
> @@ -114,6 +114,9 @@ int smc_cdc_msg_send(struct smc_connection *conn,
>  	union smc_host_cursor cfed;
>  	int rc;
>  
> +	if (unlikely(!READ_ONCE(conn->sndbuf_desc)))
> +		return -ENOBUFS;
> +
>  	smc_cdc_add_pending_send(conn, pend);
>  
>  	conn->tx_cdc_seq++;
> -- 
> 1.8.3.1

Re: [PATCH net v2] net/smc: fix NULL sndbuf_desc in smc_cdc_tx_handler()

[Wenjia Zhang]

On 08.03.23 09:17, D. Wythe wrote:
> From: "D. Wythe" <alibuda@linux.alibaba.com>
> 
> When performing a stress test on SMC-R by rmmod mlx5_ib driver
> during the wrk/nginx test, we found that there is a probability
> of triggering a panic while terminating all link groups.
> 
> This issue dues to the race between smc_smcr_terminate_all()
> and smc_buf_create().
> 
> 			smc_smcr_terminate_all
> 
> smc_buf_create
> /* init */
> conn->sndbuf_desc = NULL;
> ...
> 
> 			__smc_lgr_terminate
> 				smc_conn_kill
> 					smc_close_abort
> 						smc_cdc_get_slot_and_msg_send
> 
> 			__softirqentry_text_start
> 				smc_wr_tx_process_cqe
> 					smc_cdc_tx_handler
> 						READ(conn->sndbuf_desc->len);
> 						/* panic dues to NULL sndbuf_desc */
> 
> conn->sndbuf_desc = xxx;
> 
> This patch tries to fix the issue by always to check the sndbuf_desc
> before send any cdc msg, to make sure that no null pointer is
> seen during cqe processing.
> 
> Fixes: 0b29ec643613 ("net/smc: immediate termination for SMCR link groups")
> Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
> 

Thanks for the fix!

Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>

> v2 -> v1: change retval from EINVAL to ENOBUFS
> 
> ---
>   net/smc/smc_cdc.c | 3 +++
>   1 file changed, 3 insertions(+)
> 
> diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c
> index 53f63bf..89105e9 100644
> --- a/net/smc/smc_cdc.c
> +++ b/net/smc/smc_cdc.c
> @@ -114,6 +114,9 @@ int smc_cdc_msg_send(struct smc_connection *conn,
>   	union smc_host_cursor cfed;
>   	int rc;
>   
> +	if (unlikely(!READ_ONCE(conn->sndbuf_desc)))
> +		return -ENOBUFS;
> +
>   	smc_cdc_add_pending_send(conn, pend);
>   
>   	conn->tx_cdc_seq++;