[ovs-dev] [PATCH v2 1/3] netdev-dpdk: Fix dpdk_mp leak in case of EEXIST.

Stokes, Ian ian.stokes at intel.com
Fri Nov 10 14:14:01 UTC 2017


Hi Ilya,

Patch looks OK to me and I'm running validation before adding itto the DPDK merge branch, one nit,  would you mind re-submitting and expanding the commit message to summarize why the change is needed to keep with OVS commit message policy.

https://github.com/openvswitch/ovs/blob/master/Documentation/internals/contributing/submitting-patches.rst

Thanks
Ian
> CC: Robert Wojciechowicz <robertx.wojciechowicz at intel.com>
> CC: Antonio Fischetti <antonio.fischetti at intel.com>
> Fixes: d555d9bded5f ("netdev-dpdk: Create separate memory pool for each
> port.")
> Fixes: b6b26021d2e2 ("netdev-dpdk: fix management of pre-existing
> mempools.")
> Signed-off-by: Ilya Maximets <i.maximets at samsung.com>
> Acked-by: Antonio Fischetti <antonio.fischetti at intel.com>
> ---
>  lib/netdev-dpdk.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c index 76e79be..ad52a03
> 100644
> --- a/lib/netdev-dpdk.c
> +++ b/lib/netdev-dpdk.c
> @@ -659,6 +659,12 @@ netdev_dpdk_mempool_configure(struct netdev_dpdk
> *dev)
>           * Update dev with the new values. */
>          dev->mtu = dev->requested_mtu;
>          dev->max_packet_len = MTU_TO_FRAME_LEN(dev->mtu);
> +        /* 'mp' should contain pointer to the mempool already owned by
> netdev.
> +         * Otherwise something went completely wrong. */
> +        ovs_assert(dev->dpdk_mp);
> +        ovs_assert(dev->dpdk_mp->mp == mp->mp);
> +        /* Free the returned struct dpdk_mp because it will not be used.
> */
> +        rte_free(mp);
>          return EEXIST;
>      } else {
>          /* A new mempool was created, release the previous one. */
> --
> 2.7.4



More information about the dev mailing list