[dpdk-dev] [PATCH 1/1] ena: fix SIGFPE with 0 rx queues
Michał Krawczyk
mk at semihalf.com
Thu Jun 21 13:50:03 CEST 2018
Hi Daria,
I couldn't find the patch in the patchwork or the mailing list, so
sorry for late response.
2018-06-21 12:14 GMT+02:00 Marcin Wojtas <mw at semihalf.com>:
> + Ferruh and Michal
>
> 2018-06-20 18:32 GMT+02:00 Daria Kolistratova <daria.kolistratova at intel.com>:
>> When he number of rx queues is 0
Typo (missing 't' letter)
>> (what can be when application does not receive)
To be precise, the issue appears only when the application is
requesting ETH_MQ_RX_RSS_FLAG in the
rte_dev->data->dev_conf.rxmode.mq_mode.
>> failed with SIGFPE.
>> Fixed adding zero check before division.
>>
>> Signed-off-by: Daria Kolistratova <daria.kolistratova at intel.com>
>> ---
>> drivers/net/ena/ena_ethdev.c | 6 +++++-
>> 1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
>> index 9ae73e331..76c483921 100644
>> --- a/drivers/net/ena/ena_ethdev.c
>> +++ b/drivers/net/ena/ena_ethdev.c
>> @@ -684,7 +684,11 @@ static int ena_rss_init_default(struct ena_adapter *adapter)
>> }
>>
>> for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) {
>> - val = i % nb_rx_queues;
>> + if (nb_rx_queues != 0)
>> + val = i % nb_rx_queues;
>> + else
>> + val = 0;
>> +
We can skip the RSS configuration at all if the nb_rx_queues is 0. I
suggest to modify the patch like this:
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index 8701328ba..5ba90f58a 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -1056,7 +1056,7 @@ static int ena_start(struct rte_eth_dev *dev)
return rc;
if (adapter->rte_dev->data->dev_conf.rxmode.mq_mode &
- ETH_MQ_RX_RSS_FLAG) {
+ ETH_MQ_RX_RSS_FLAG && adapter->rte_dev->data->nb_rx_queues > 0) {
rc = ena_rss_init_default(adapter);
if (rc)
return rc;
>> rc = ena_com_indirect_table_fill_entry(ena_dev, i,
>> ENA_IO_RXQ_IDX(val));
>> if (unlikely(rc && (rc != ENA_COM_UNSUPPORTED))) {
>> --
>> 2.14.4
>>
More information about the dev
mailing list