[dpdk-test-report] | ERROR | daily Intel builds (53/55)
Ferruh Yigit
ferruh.yigit at intel.com
Mon Oct 22 12:46:26 CEST 2018
On 10/21/2018 9:30 PM, sys_stv at intel.com wrote:
> Last commit Time : Fri Oct 19 22:37:10 2018 +0200
> Last Author : Thomas Monjalon
> Last commit Hash : 739e13bcc98f562d3301f808ec76507ebae82e63
> Last comment : devargs: fix freeing during device removal
>
> Build Summary : 55 Builds Done, 53 Successful, 2 Failures.
>
> Failure details :
>
> Failure #1
> FC18_64 / Linux 3.6.10-4 / GCC 4.7.2
> Config: x86_64-native-linuxapp-gcc
> CONFIG_RTE_LIBRTE_PMD_PCAP=y,
> CONFIG_RTE_NIC_BYPASS=y,
> CONFIG_RTE_BUILD_SHARED_LIB=y,
> CONFIG_RTE_LIBRTE_VHOST=y,
> CONFIG_RTE_LIBRTE_PMD_QAT=y,
> CONFIG_RTE_LIBRTE_PMD_AESNI_GCM=y,
> CONFIG_RTE_LIBRTE_PMD_NULL_CRYPTO=y
>
> DPDK/drivers/net/tap/rte_eth_tap.c: In function ‘tun_alloc’:
> DPDK/drivers/net/tap/rte_eth_tap.c:176:29: error: ‘IFF_DETACH_QUEUE’ undeclared (first use in this function)
> compilation terminated due to -Wfatal-errors.
>
> Failure #2
> FC18_64 / Linux 3.6.10-4 / ICC 16.0.2
> Config: x86_64-native-linuxapp-icc
> CONFIG_RTE_LIBRTE_PMD_PCAP=y,
> CONFIG_RTE_NIC_BYPASS=y,
> CONFIG_RTE_BUILD_SHARED_LIB=y,
> CONFIG_RTE_LIBRTE_VHOST=y,
> CONFIG_RTE_LIBRTE_PMD_QAT=y,
> CONFIG_RTE_LIBRTE_PMD_AESNI_GCM=y,
> CONFIG_RTE_LIBRTE_PMD_NULL_CRYPTO=y
>
> DPDK/drivers/net/tap/rte_eth_tap.c: In function ‘tun_alloc’:
> DPDK/drivers/net/tap/rte_eth_tap.c:176:29: error: ‘IFF_DETACH_QUEUE’ undeclared (first use in this function)
> compilation terminated due to -Wfatal-errors.
This is because older kernels (< 3.8) doesn't have ‘IFF_DETACH_QUEUE’ defined.
Fedora 18 EOL on beginning of the 2014, should we still include it on our test
harness? Any objection to remove it?
And it seems Fedora coverage below is like:
FC18, FC20, FC21, FC24, FC25.
1- Why are the gaps for, intentional?
2- Why we don't have latest ones, FC27 & FC28 ?
[1]
https://en.wikipedia.org/wiki/Fedora_version_history
More information about the test-report
mailing list