Please use latest code from git, libnuma dependency has been removed there as 
you noticed, and upgrade gcc (my ubuntu has version 4.8.4).

Alfredo

> On 19 Oct 2015, at 08:59, Bryan Fok <[email protected]> wrote:
> 
> I did a lot of testing and encountered multiple problems.
> 
> 1) my other computer has ZC and seems to be working, and it print out is 
> license 6.1.1.150917. the computer which is not working having license 6.0.x.
> 
> 
> 2) so I try to install 6.1.1 on it. I copy the same library over and get the 
> libnuma error at .userland/lib/configure. 
> 
> NOTE: In fact I had libnuma-dev installed.
> ii  libnuma-dbg                                 2.0.8~rc3-1                   
>         Debug package for libnuma
> ii  libnuma-dev                                 2.0.8~rc3-1                   
>         Development files for libnuma
> ii  libnuma1                                    2.0.8~rc3-1                   
>         Libraries for controlling NUMA policy
> ii  numactl                                     2.0.8~rc3-1   
>  
> ./configure output
> ....
> hecking librdi.h usability... no
> checking librdi.h presence... no
> checking for librdi.h... no
> checking for pthread_setaffinity_np in -lpthread... no
> checking for librdi.h... (cached) no
> checking for redisCommand in -lhiredis... no
> checking for numa_available in -lnuma... no
> checking if libnuma is present... no
> Please install libnuma-dev in order to use PF_RING
> 
>  
> 3) and than I tried the new dev download from  
> <https://github.com/ntop/PF_RING:> 
> <https://github.com/ntop/PF_RING,>https://github.com/ntop/PF_RING 
> <https://github.com/ntop/PF_RING> and i get this:
> 
> checking librdi.h usability... no
> checking librdi.h presence... no
> checking for librdi.h... no
> checking for pthread_setaffinity_np in -lpthread... no
> checking for librdi.h... (cached) no
> checking for redisCommand in -lhiredis... no
> checking for clock_gettime in -lrt... no
> checking for nl_handle_alloc in -lnl... no
> checking for dag_get_stream_erf_types in -ldag... no
> checking for NT_Init in -lntapi... no
> checking if ntapi is present... checking Accolade support... no
> checking Myricom support... no
> configure: creating ./config.status
> config.status: creating Makefile
> config.status: creating pfring_config
> config.status: creating ../examples/Makefile
> config.status: creating ../examples_zc/Makefile
> config.status: creating ../examples_libzero/Makefile
> config.status: creating ../c++/Makefile
> config.status: creating config.h
> config.status: config.h is unchanged
> make[2]: Entering directory `/home/bfok/Downloads/PF_RING-dev/userland/lib'
> gcc -march=core-avx2 -mtune=core-avx2  -Wall -fPIC -I../../kernel 
> -I../libpcap          -D ENABLE_BPF  -D ENABLE_HW_TIMESTAMP   -O2  -c 
> pfring.c -o pfring.o
> pfring.c:1:0: error: bad value (core-avx2) for -march= switch
> pfring.c:1:0: error: bad value (core-avx2) for -mtune= switch
> make[2]: *** [pfring.o] Error 1
> make[2]: Leaving directory `/home/bfok/Downloads/PF_RING-dev/userland/lib'
> make[1]: *** [libpfring] Error 2
> make[1]: Leaving directory `/home/bfok/Downloads/PF_RING-dev/userland'
> make: *** [all] Error 2
> 
> 
> 4) the funny thing is the other computer which installed 6.1.1 can compile 
> this new pf_ring version, and they have exactly the same version of ubuntu 
> (Ubuntu 12.04.5 LTS), and gcc (gcc version 4.6.3 (Ubuntu/Linaro 
> 4.6.3-1ubuntu5) ).
> 
> Bryan
> 
>  
> 
> 
> 
> From: [email protected]
> Date: Fri, 16 Oct 2015 15:00:50 +0200
> To: [email protected]
> Subject: Re: [Ntop-misc] Failed to send packet total size greater than        
> 64bytes using code from the zsend example
> 
> Hi Brian
> it does work for me with the same configuration
> 
> # ethtool -i eth0
> driver: e1000e
> version: 3.2.4.2-NAPI
> ..
> 
> # ./zbalance_ipc -i zc:eth0 -c 1 -n 1
> max_packet_len=1536
> ..
> 
> Alfredo
> 
> On 16 Oct 2015, at 12:10, Bryan Fok <[email protected] 
> <mailto:[email protected]>> wrote:
> 
> The driver i am using is under ZC e1000e FYI
> 
> From: [email protected] <mailto:[email protected]>
> To: [email protected] <mailto:[email protected]>
> Date: Fri, 16 Oct 2015 09:52:14 +0000
> Subject: Re: [Ntop-misc] Failed to send packet total size greater than 
> 64bytes using code from the zsend example
> 
> BTW I am passing zc:eth1 as the device_name argument
> 
> From: [email protected] <mailto:[email protected]>
> To: [email protected] <mailto:[email protected]>
> Date: Fri, 16 Oct 2015 09:49:23 +0000
> Subject: Re: [Ntop-misc] Failed to send packet total size greater than 
> 64bytes using code from the zsend example
> 
> Thank you Alfredo
> 
> both print 0. How do I fix this? 
> 
> Bryan
> 
> From: [email protected] <mailto:[email protected]>
> Date: Fri, 16 Oct 2015 10:01:33 +0200
> To: [email protected] <mailto:[email protected]>
> Subject: Re: [Ntop-misc] Failed to send packet total size greater than        
> 64bytes using code from the zsend example
> 
> Please print:
> 1. asio::pf_ring::max_packet_len(device_name.c_str())
> 2. buffers[0]->len before pfring_zc_send_pkt()
> 
> Alfredo
> 
> On 16 Oct 2015, at 08:52, Bryan Fok <[email protected] 
> <mailto:[email protected]>> wrote:
> 
> FYI non zero-copy does not have this limit
> 
> From: [email protected] <mailto:[email protected]>
> To: [email protected] <mailto:[email protected]>
> Date: Fri, 16 Oct 2015 06:50:32 +0000
> Subject: [Ntop-misc] Failed to send packet total size greater than 64bytes 
> using code from the zsend example
> 
> pfring_zc_send_pkt() always return -1 when the message size (include the 
> header + my application data) is greater than 64 bytes. Please help.
> 
> 
> This is how I create the cluster
> 
> zc = pfring_zc_create_cluster(cluster_id, 
> asio::pf_ring::max_packet_len(device_name.c_str()), 0, NBUFF, 
> numa_node_of_cpu(bind_cpu), NULL /* auto hugetlb mountpoint */
>         );
> 
>         if (zc == NULL)
>         {
>             fprintf(stderr, "pfring_zc_create_cluster error [%s] Please check 
> your hugetlb configuration\n", strerror(errno));
>             return -1;
>         }
> 
>         zq = pfring_zc_open_device(zc, device_name.c_str(), tx_only, 0);
> 
>         if (zq == NULL)
>         {
>             fprintf(stderr, "pfring_zc_open_device error [%s] Please check 
> that %s is up and not already used\n", strerror(errno),
>                     device_name.c_str());
>             return -1;
>         }
> 
>         fprintf(stderr, "Sending packets to %s\n", device_name.c_str());
> 
>         for (int i = 0; i < NBUFF; i++)
>         {
>             buffers[i] = pfring_zc_get_packet_handle(zc);
> 
>             if (buffers[i] == NULL)
>             {
>                 fprintf(stderr, "pfring_zc_get_packet_handle error\n");
>                 return -1;
>             }
>         }
> 
> 
> This is how I send the packet
> 
>     char a[23] = {0}; //23 + 42 = 65 return -1 from the send function, 
> resetting a to 22 it works
>     int tot_msg_size = forge_udp_packet(pfring_zc_pkt_buff_data(buffers[0], 
> zq), 1600, a, sizeof(a), parse_ip_v4_string(sender_ip.c_str()),
>                     sender_port, parse_ip_v4_string(mcast_ip.c_str()), 
> mcast_port);
> 
>     std::cout << "tot_msg_size:" << std::dec << tot_msg_size << std::endl;
> 
>     buffers[0]->len = tot_msg_size;
>     std::cout << "buffers[0] : ";
>     for (int i = 0; i < buffers[0]->len; ++i)
>     {
>           std::cout << std::hex << (int) pfring_zc_pkt_buff_data(buffers[0], 
> zq)[i] << " ";
>     }
>     int rc = pfring_zc_send_pkt(zq, &buffers[0], 0);
> 
> 
> B.R
> Bryan
> 
> _______________________________________________ Ntop-misc mailing list 
> [email protected] <mailto:[email protected]> 
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc 
> <http://listgateway.unipi.it/mailman/listinfo/ntop-misc>_______________________________________________
> Ntop-misc mailing list
> [email protected] <mailto:[email protected]>
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc 
> <http://listgateway.unipi.it/mailman/listinfo/ntop-misc>
> 
> _______________________________________________ Ntop-misc mailing list 
> [email protected] <mailto:[email protected]> 
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc 
> <http://listgateway.unipi.it/mailman/listinfo/ntop-misc>
> _______________________________________________ Ntop-misc mailing list 
> [email protected] <mailto:[email protected]> 
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc 
> <http://listgateway.unipi.it/mailman/listinfo/ntop-misc>
> _______________________________________________ Ntop-misc mailing list 
> [email protected] <mailto:[email protected]> 
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc 
> <http://listgateway.unipi.it/mailman/listinfo/ntop-misc>_______________________________________________
> Ntop-misc mailing list
> [email protected] <mailto:[email protected]>
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc 
> <http://listgateway.unipi.it/mailman/listinfo/ntop-misc>
> 
> _______________________________________________ Ntop-misc mailing list 
> [email protected] <mailto:[email protected]> 
> http://listgateway.unipi.it/mailman/listinfo/ntop-miscg 
> <http://listgateway.unipi.it/mailman/listinfo/ntop-miscg>_______________________________________________
> Ntop-misc mailing list
> [email protected] <mailto:[email protected]>
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc 
> <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