Hi Marco
please pull latest code from git, it seems you are running an old script which 
is loading multiple RSS queues, and you are capturing just from the first one.

Alfredo

> On 08 Jun 2015, at 09:30, Marco GarrĂ© <[email protected]> wrote:
> 
> Hi,
> 
> I'm trying to use PF_RING in ZC mode with intel X710, but don't seems work 
> fine.
> My server is:
>    3.16.0-4-amd64 #1 SMP Debian 3.16.7-ckt9-3 (2015-04-23) x86_64 GNU/Linux
>    Ethernet controller: Intel Corporation Ethernet Controller X710 for 10GbE 
> SFP+ (rev 01)
>    PF_RING_6.0.3
> 
> When I run "drivers/ZC/intel/i40e/i40e-1.1.23/src/load_driver.sh" in  the 
> result is:
> 
> Warning: 512 hugepages available, 1024 requested
> find: `/sys/kernel/debug/i40e/': No such file or directory
> find: `/sys/kernel/debug/i40e/': No such file or directory
> irqbalance: no process found
> Configuring eth12
> no rx vectors found on eth12
> no tx vectors found on eth12
> eth12 mask=1 for /proc/irq/147/smp_affinity
> eth12 mask=2 for /proc/irq/148/smp_affinity
> ...
> eth12 mask=400000 for /proc/irq/196/smp_affinity
> eth12 mask=800000 for /proc/irq/197/smp_affinity
> Configuring eth2
> no rx vectors found on eth2
> no tx vectors found on eth2
> eth2 mask=1 for /proc/irq/208/smp_affinity
> eth2 mask=2 for /proc/irq/209/smp_affinity
> ...
> eth2 mask=400000 for /proc/irq/248/smp_affinity
> eth2 mask=800000 for /proc/irq/249/smp_affinity
> Configuring eth3
> no rx vectors found on eth3
> no tx vectors found on eth3
> eth3 mask=1 for /proc/irq/260/smp_affinity
> eth3 mask=2 for /proc/irq/261/smp_affinity
> ...
> eth3 mask=400000 for /proc/irq/282/smp_affinity
> eth3 mask=800000 for /proc/irq/283/smp_affinity
> Configuring eth13
> no rx vectors found on eth13
> no tx vectors found on eth13
> eth13 mask=1 for /proc/irq/113/smp_affinity
> eth13 mask=2 for /proc/irq/114/smp_affinity
> ...
> eth13 mask=400000 for /proc/irq/135/smp_affinity
> eth13 mask=800000 for /proc/irq/136/smp_affinity
> 
> if I list loaded modules I can see i40e and pf_ring.
> 
> When I try to use the example code zsend and zcount in ZC mode, the sender 
> works fine, but the counter drop 90% of packets.
> For example if zsend have this stats: Actual Stats: 14'882'590.15 pps - 10.00 
> Gbps
> zcount have: Actual Stats: 540'820.30 pps (14'340'221.54 drops) - 0.36 Gbps
> 
> If i try to use NO ZC mode, it seems not lose a lots of packets.
> For example if zsend have this stats: Actual Stats: 681'143.93 pps - 0.46 Gbps
> zcount have: Actual Stats: 659'597.03 pps (0.00 drops) - 0.44 Gbps
> 
> Could you have any idea for fix the ZC mode behaviour ?
> 
> Thanks for any responses.
> 
> Marco.
> _______________________________________________
> Ntop-misc mailing list
> [email protected]
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc

_______________________________________________
Ntop-misc mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop-misc

Reply via email to