[PATCH] app/testpmd: fix crash in multi-process packet forwarding

[fengchengwen]

Hi Dengdui,

On 2024/1/26 10:41, Dengdui Huang wrote:
> On multi-process scenario, each process creates flows based on the
> number of queues. When nbcore is greater than 1, multiple cores may
> use the same queue to forward packet, like:
> dpdk-testpmd -a BDF --proc-type=auto -- -i --rxq=4 --txq=4
> --nb-cores=2 --num-procs=2 --proc-id=0
> testpmd> start
> mac packet forwarding - ports=1 - cores=2 - streams=4 - NUMA support
> enabled, MP allocation mode: native
> Logical Core 2 (socket 0) forwards packets on 2 streams:
> RX P=0/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
> RX P=0/Q=1 (socket 0) -> TX P=0/Q=1 (socket 0) peer=02:00:00:00:00:00
> Logical Core 3 (socket 0) forwards packets on 2 streams:
> RX P=0/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
> RX P=0/Q=1 (socket 0) -> TX P=0/Q=1 (socket 0) peer=02:00:00:00:00:00

tip: it would be more readable if with an indent, just like below example.

Acked-by: Chengwen Feng <fengchengwen at huawei.com>

Thanks

> 
> After this commit, the result will be:
> dpdk-testpmd -a BDF --proc-type=auto -- -i --rxq=4 --txq=4
> --nb-cores=2 --num-procs=2 --proc-id=0
> testpmd> start
> io packet forwarding - ports=1 - cores=2 - streams=2 - NUMA support
> enabled, MP allocation mode: native
> Logical Core 2 (socket 0) forwards packets on 1 streams:
>   RX P=0/Q=0 (socket 2) -> TX P=0/Q=0 (socket 2) peer=02:00:00:00:00:00
> Logical Core 3 (socket 0) forwards packets on 1 streams:
>   RX P=0/Q=1 (socket 2) -> TX P=0/Q=1 (socket 2) peer=02:00:00:00:00:00
> 
> Fixes: a550baf24af9 ("app/testpmd: support multi-process")
> Cc: stable at dpdk.org
> 
> Signed-off-by: Dengdui Huang <huangdengdui at huawei.com>
> ---
>  app/test-pmd/config.c | 6 +-----
>  1 file changed, 1 insertion(+), 5 deletions(-)
> 
> diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
> index cad7537bc6..2c4dedd603 100644
> --- a/app/test-pmd/config.c
> +++ b/app/test-pmd/config.c
> @@ -4794,7 +4794,6 @@ rss_fwd_config_setup(void)
>  	queueid_t  nb_q;
>  	streamid_t  sm_id;
>  	int start;
> -	int end;
>  
>  	nb_q = nb_rxq;
>  	if (nb_q > nb_txq)
> @@ -4802,7 +4801,7 @@ rss_fwd_config_setup(void)
>  	cur_fwd_config.nb_fwd_lcores = (lcoreid_t) nb_fwd_lcores;
>  	cur_fwd_config.nb_fwd_ports = nb_fwd_ports;
>  	cur_fwd_config.nb_fwd_streams =
> -		(streamid_t) (nb_q * cur_fwd_config.nb_fwd_ports);
> +		(streamid_t) (nb_q / num_procs * cur_fwd_config.nb_fwd_ports);
>  
>  	if (cur_fwd_config.nb_fwd_streams < cur_fwd_config.nb_fwd_lcores)
>  		cur_fwd_config.nb_fwd_lcores =
> @@ -4824,7 +4823,6 @@ rss_fwd_config_setup(void)
>  	 * the 2~3 queue for secondary process.
>  	 */
>  	start = proc_id * nb_q / num_procs;
> -	end = start + nb_q / num_procs;
>  	rxp = 0;
>  	rxq = start;
>  	for (sm_id = 0; sm_id < cur_fwd_config.nb_fwd_streams; sm_id++) {
> @@ -4843,8 +4841,6 @@ rss_fwd_config_setup(void)
>  			continue;
>  		rxp = 0;
>  		rxq++;
> -		if (rxq >= end)
> -			rxq = start;
>  	}
>  }
>  
>