[dpdk-dev,2/2] net/ixgbe: fix l3fwd start failed on PF

Message ID 20171102170432.168194-2-yanglong.wu@intel.com (mailing list archive)
State Superseded, archived
Delegated to: Ferruh Yigit
Headers

Checks

Context Check Description
ci/checkpatch warning coding style issues
ci/Intel-compilation success Compilation OK

Commit Message

Yanglong Wu Nov. 2, 2017, 5:04 p.m. UTC
  which occurred when the SRIOV is active and tx_q > rx_q.
The number of nb_q_per_pool should equeal to max number
of queues supported by HW not nb_rx_q.

Fixes: 27b609cbd1c6 (ethdev: move the multi-queue mode 
check to specific drivers)

Signed-off-by: Yanglong Wu <yanglong.wu@intel.com>
---
 drivers/net/ixgbe/ixgbe_ethdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
  

Comments

Ananyev, Konstantin Nov. 2, 2017, 2:05 p.m. UTC | #1
Hi,

> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Yanglong Wu
> Sent: Thursday, November 2, 2017 5:05 PM
> To: dev@dpdk.org
> Cc: Wu, Yanglong <yanglong.wu@intel.com>
> Subject: [dpdk-dev] [PATCH 2/2] net/ixgbe: fix l3fwd start failed on PF
> 
> which occurred when the SRIOV is active and tx_q > rx_q.
> The number of nb_q_per_pool should equeal to max number
> of queues supported by HW not nb_rx_q.
> 
> Fixes: 27b609cbd1c6 (ethdev: move the multi-queue mode
> check to specific drivers)
> 
> Signed-off-by: Yanglong Wu <yanglong.wu@intel.com>
> ---
>  drivers/net/ixgbe/ixgbe_ethdev.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
> index ae9c44421..0f0641da1 100644
> --- a/drivers/net/ixgbe/ixgbe_ethdev.c
> +++ b/drivers/net/ixgbe/ixgbe_ethdev.c
> @@ -2180,7 +2180,7 @@ ixgbe_check_vf_rss_rxq_num(struct rte_eth_dev *dev, uint16_t nb_rx_q)
>  		return -EINVAL;
>  	}
> 
> -	RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = nb_rx_q;
> +	RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = 128/RTE_ETH_DEV_SRIOV(dev).active;
>  	RTE_ETH_DEV_SRIOV(dev).def_pool_q_idx = pci_dev->max_vfs * nb_rx_q;
> 
>  	return 0;
> --
> 2.11.0

Not sure I understand what is the purpose of that patch...
Do you want to prevent RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = 1?  
Konstantin
  
Jingjing Wu Nov. 7, 2017, 8:39 a.m. UTC | #2
> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Ananyev, Konstantin
> Sent: Thursday, November 2, 2017 10:06 PM
> To: Wu, Yanglong <yanglong.wu@intel.com>; dev@dpdk.org
> Cc: Wu, Yanglong <yanglong.wu@intel.com>
> Subject: Re: [dpdk-dev] [PATCH 2/2] net/ixgbe: fix l3fwd start failed on PF
> 
> Hi,
> 
> > -----Original Message-----
> > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Yanglong Wu
> > Sent: Thursday, November 2, 2017 5:05 PM
> > To: dev@dpdk.org
> > Cc: Wu, Yanglong <yanglong.wu@intel.com>
> > Subject: [dpdk-dev] [PATCH 2/2] net/ixgbe: fix l3fwd start failed on PF
> >
> > which occurred when the SRIOV is active and tx_q > rx_q.
> > The number of nb_q_per_pool should equeal to max number
> > of queues supported by HW not nb_rx_q.
> >
> > Fixes: 27b609cbd1c6 (ethdev: move the multi-queue mode
> > check to specific drivers)
> >
> > Signed-off-by: Yanglong Wu <yanglong.wu@intel.com>
> > ---
> >  drivers/net/ixgbe/ixgbe_ethdev.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
> > index ae9c44421..0f0641da1 100644
> > --- a/drivers/net/ixgbe/ixgbe_ethdev.c
> > +++ b/drivers/net/ixgbe/ixgbe_ethdev.c
> > @@ -2180,7 +2180,7 @@ ixgbe_check_vf_rss_rxq_num(struct rte_eth_dev *dev,
> uint16_t nb_rx_q)
> >  		return -EINVAL;
> >  	}
> >
> > -	RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = nb_rx_q;
> > +	RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = 128/RTE_ETH_DEV_SRIOV(dev).active;
> >  	RTE_ETH_DEV_SRIOV(dev).def_pool_q_idx = pci_dev->max_vfs * nb_rx_q;
> >
> >  	return 0;
> > --
> > 2.11.0
> 
> Not sure I understand what is the purpose of that patch...
> Do you want to prevent RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = 1?
> Konstantin
> 
I think his purpose is to set the RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool to be the max number of queues in one pool
according to the how to split the queue index.

Now, for rss and virtualization mode, ixgbe has combination like 2 queues * 64 pools and 4 queues * 32 pools.

BTW, I think the title of this patch need to be reword. It looks confusing if it is a bug in ixgbe.
> 
> 
>
  

Patch

diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index ae9c44421..0f0641da1 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -2180,7 +2180,7 @@  ixgbe_check_vf_rss_rxq_num(struct rte_eth_dev *dev, uint16_t nb_rx_q)
 		return -EINVAL;
 	}
 
-	RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = nb_rx_q;
+	RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = 128/RTE_ETH_DEV_SRIOV(dev).active;
 	RTE_ETH_DEV_SRIOV(dev).def_pool_q_idx = pci_dev->max_vfs * nb_rx_q;
 
 	return 0;