Re: [ovs-discuss] OVN Failed Flow Offload

2023-02-06 Thread David Marchand via discuss
el. I don't see why some ethernet address would require some special offloading considerations, but maybe others have a better opinion on this topic. -- David Marchand ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailm

Re: [ovs-discuss] OVN Failed Flow Offload

2023-02-06 Thread David Marchand via discuss
rop / end > 2023-02-06T15:04:03.588Z|00136|dpdk(hw_offload105)|ERR|bnxt_validate_and_parse_flow_type(): > DMAC is invalid! This flow (for lldp frames) is rejected by the DPDK net/bnxt driver. Copying this driver MAINTAINERS, maybe they can give the reason why this driver won't accept su

Re: [ovs-discuss] DPDK issues with OVS, help?

2022-07-19 Thread David Marchand
re different, and must be left bound to their respective mlx kernel drivers. $ dpdk-pmdinfo.py $(which ovs-vswitchd) | grep -A1 'PMD NAME: mlx5' PMD NAME: mlx5_common_pci PMD NAME: mlx5_common_driver -- PMD NAME: mlx5_eth PMD KMOD DEPENDENCIES: * ib_uverbs

Re: [ovs-discuss] DPDK issues with OVS, help?

2022-07-18 Thread David Marchand
ncy cache): # meson --reconfigure build ... Library numa found: YES Has header "numaif.h" : YES ... $ grep RTE_HAS_LIBNUMA build/rte_build_config.h #define RTE_HAS_LIBNUMA 1 $ grep RTE_EAL_NUMA_AWARE_HUGEPAGES build/rte_build_config.h #define RTE_EAL_NUMA_AWARE_HUGEPAGES 1 -- David Marc

Re: [ovs-discuss] DPDK issues with OVS, help?

2022-07-18 Thread David Marchand
DK expects hugepage memory. >From this error message (which comes with a DPDK init failure, not a crash), I suspect you don't have enough hugepages declared. The DPDK (and OVS) documentation has examples of how to configure this. Like for example: https://github.com/openvswitch/ovs/blob/master/Documentation/intro/install/dpdk.rst -- David Marchand ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss

Re: [ovs-discuss] OVS DPDK KNI support

2022-04-04 Thread David Marchand
ype=dpdk ofport_request=2 -- set Interface vEth0 options:dpdk-devargs=net_kni0 -- David Marchand ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss

Re: [ovs-discuss] OVS (tag: 2.16.2) Build Fails With DPDK 21.11

2022-01-18 Thread David Marchand
s to link against DPDK 21.11, since DPDK_21 ABI has not been preserved. -- David Marchand ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss

Re: [ovs-discuss] how to make ovs2.13 with dpdk19.11(which is compiled by meson)

2021-12-07 Thread David Marchand
fig that comes with RHEL/Centos 7. https://github.com/openvswitch/ovs/commit/252e1e576443#diff-b781fde347898f87c1814ea8961b8e39c2615374b2811e7fc91047e7517d9b57R356 -- David Marchand ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss

Re: [ovs-discuss] Intel X550T2 nic dpdk issue

2021-10-19 Thread David Marchand
fails or max number of > > port is reached. > > 'Invalid port_id=32' goes from DPDK. OVS has no such logs. > By this, I meant that this log is displayed because of how OVS probing code reports a DPDK_ETH_PORT_ID_INVALID on DPDK probe error. Later, OVS asks dpdk about DPDK_ETH_PO

Re: [ovs-discuss] Intel X550T2 nic dpdk issue

2021-10-19 Thread David Marchand
ian and Luca in the loop. - About log messages above, what might be confusing for final users is that DPDK won't log a message when no driver is available. One reason it is done like this, is to avoid spamming messages for every pci devices on the system at PCI scan time. Maybe DPDK could do bette