Re: [dpdk-dev] [PATCH v3 3/6] app/testpmd: remove restriction on txpkts set

["Wei Hu (Xavier)" <huwei013@chinasoftinc.com>]

Hi, Ferruh Yigit

On 2020/9/22 22:51, Ferruh Yigit wrote:
> On 9/19/2020 11:47 AM, Wei Hu (Xavier) wrote:
>> From: Chengchang Tang <tangchengchang@huawei.com>
>>
>> Currently, if nb_txd is not set, the txpkts is not allowed to be set
>> because the nb_txd is used to avoid the numer of segments exceed the Tx
>> ring size and the default value of nb_txd is 0. And there is a bug that
>> nb_txd is the global configuration for Tx ring size and the ring size
>> could be changed by some command per queue. So these valid check is
>> unreliable and introduced unnecessary constraints.
>>
>> This patch adds a valid check function to use the real Tx ring size to
>> check the validity of txpkts.
>>
>> Fixes: af75078fece3 ("first public release")
>> Cc: stable@dpdk.org
>>
>> Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
>> Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
>> ---
>>   app/test-pmd/config.c | 42 ++++++++++++++++++++++++++++++++++++++----
>>   1 file changed, 38 insertions(+), 4 deletions(-)
>>
>> diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
>> index 4e33208..882de2d 100644
>> --- a/app/test-pmd/config.c
>> +++ b/app/test-pmd/config.c
>> @@ -2984,17 +2984,51 @@ show_tx_pkt_segments(void)
>>       printf("Split packet: %s\n", split);
>>   }
>>   +static bool
>> +nb_segs_is_invalid(unsigned int nb_segs)
>> +{
>> +    uint16_t port_id;
>> +
>> +    RTE_ETH_FOREACH_DEV(port_id) {
>> +        struct rte_port *port = &ports[port_id];
>> +        uint16_t ring_size;
>> +        uint16_t queue_id;
>> +
>> +        /*
>> +         * When configure the txq by rte_eth_tx_queue_setup with
>> +         * nb_tx_desc being 0, it will use a default value provided by
>> +         * PMDs to setup this txq. If the default value is 0, it will
>> +         * use the RTE_ETH_DEV_FALLBACK_TX_RINGSIZE to setup this txq.
>> +         */
>> +        for (queue_id = 0; queue_id < nb_txq; queue_id++) {
>> +            if (port->nb_tx_desc[queue_id])
>> +                ring_size = port->nb_tx_desc[queue_id];
>> +            else if (port->dev_info.default_txportconf.ring_size)
>> +                ring_size =
>> + port->dev_info.default_txportconf.ring_size;
>> +            else
>> +                ring_size = RTE_ETH_DEV_FALLBACK_TX_RINGSIZE;
>> +
>> +            if (ring_size < nb_segs) {
>> +                printf("nb segments per TX packets=%u >= TX "
>> +                       "queue(%u) ring_size=%u - ignored\n",
>> +                    nb_segs, queue_id, ring_size);
>> +                return true;
>> +            }
>> +        }
>
> What do you think calling 'rte_eth_rx_queue_info_get()' & 
> 'rte_eth_tx_queue_info_get()' to get the 'nb_desc'?

Currently not all PMD driver implement the .rxq_info_get and

.txq_info_get hook function. If calling rte_eth_rx_queue_info_get

return -ENOSTUP, we still need to obtain the ring_size in this way.


Thanks, Xavier