* [dpdk-dev] [PATCH] net/avp: remove resources when port is closed
@ 2019-05-27 13:46 Allain Legacy
2019-05-27 16:03 ` Ferruh Yigit
0 siblings, 1 reply; 2+ messages in thread
From: Allain Legacy @ 2019-05-27 13:46 UTC (permalink / raw)
To: thomas; +Cc: dev, ferruh.yigit, Matt Peters
The rte_eth_dev_close() function now handles freeing resources for
devices (e.g., mac_addrs). There is no change in behaviour for AVP
devices since they do not currently free their dynamic memory; which was
a bug. To conform with the new close() behaviour we are asserting the
RTE_ETH_DEV_CLOSE_REMOVE flag so that rte_eth_dev_close() releases
all device related dynamic memory.
Cc: Matt Peters <matt.peters@windriver.com>
Signed-off-by: Allain Legacy <allain.legacy@windriver.com>
---
drivers/net/avp/avp_ethdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/avp/avp_ethdev.c b/drivers/net/avp/avp_ethdev.c
index 09388d05f..7c37f4c8a 100644
--- a/drivers/net/avp/avp_ethdev.c
+++ b/drivers/net/avp/avp_ethdev.c
@@ -959,6 +959,8 @@ eth_avp_dev_init(struct rte_eth_dev *eth_dev)
eth_dev->dev_ops = &avp_eth_dev_ops;
eth_dev->rx_pkt_burst = &avp_recv_pkts;
eth_dev->tx_pkt_burst = &avp_xmit_pkts;
+ /* Let rte_eth_dev_close() release the port resources */
+ eth_dev->data->dev_flags |= RTE_ETH_DEV_CLOSE_REMOVE;
if (rte_eal_process_type() != RTE_PROC_PRIMARY) {
/*
--
2.12.1
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [dpdk-dev] [PATCH] net/avp: remove resources when port is closed
2019-05-27 13:46 [dpdk-dev] [PATCH] net/avp: remove resources when port is closed Allain Legacy
@ 2019-05-27 16:03 ` Ferruh Yigit
0 siblings, 0 replies; 2+ messages in thread
From: Ferruh Yigit @ 2019-05-27 16:03 UTC (permalink / raw)
To: Allain Legacy, thomas; +Cc: dev, Matt Peters
On 5/27/2019 2:46 PM, Allain Legacy wrote:
> The rte_eth_dev_close() function now handles freeing resources for
> devices (e.g., mac_addrs). There is no change in behaviour for AVP
> devices since they do not currently free their dynamic memory; which was
> a bug. To conform with the new close() behaviour we are asserting the
> RTE_ETH_DEV_CLOSE_REMOVE flag so that rte_eth_dev_close() releases
> all device related dynamic memory.
>
> Cc: Matt Peters <matt.peters@windriver.com>
> Signed-off-by: Allain Legacy <allain.legacy@windriver.com>
+1 to patch.
But Rx/Tx queues are not freed by 'rte_eth_dev_close()', they are still need to
be freed by the driver.
Since you started this, can you please extend the patch to clean them as well?
> ---
> drivers/net/avp/avp_ethdev.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/net/avp/avp_ethdev.c b/drivers/net/avp/avp_ethdev.c
> index 09388d05f..7c37f4c8a 100644
> --- a/drivers/net/avp/avp_ethdev.c
> +++ b/drivers/net/avp/avp_ethdev.c
> @@ -959,6 +959,8 @@ eth_avp_dev_init(struct rte_eth_dev *eth_dev)
> eth_dev->dev_ops = &avp_eth_dev_ops;
> eth_dev->rx_pkt_burst = &avp_recv_pkts;
> eth_dev->tx_pkt_burst = &avp_xmit_pkts;
> + /* Let rte_eth_dev_close() release the port resources */
> + eth_dev->data->dev_flags |= RTE_ETH_DEV_CLOSE_REMOVE;
>
> if (rte_eal_process_type() != RTE_PROC_PRIMARY) {
> /*
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2019-05-27 16:03 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-05-27 13:46 [dpdk-dev] [PATCH] net/avp: remove resources when port is closed Allain Legacy
2019-05-27 16:03 ` Ferruh Yigit
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).