[dpdk-dev] Meaning of ETH_TXQ_FLAGS_NOREFCOUNT

Hello, I would have thought that ETH_TXQ_FLAGS_NOREFCOUNT, when enabled, would mean to ignore ref counts, as commented in rte_ethdev.h: #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */ But it seems to be used differently: if (!(txq->txq_flags & (uint32_t)ETH_TXQ_FLAGS_N

[dpdk-dev] rte_eth_dev_socket_id() vs KVM/AWS/...

Hello Dpdk, rte_eth_dev_socket_id() describes a -1 return value as: *Returns* The NUMA socket id to which the Ethernet device is connected or a default of zero if the socket could not be determined. -1 is returned is the port_id value is out of range. But, rte_eth_dev_socket_id() is implemented

Re: [dpdk-dev] [PATCH 1/7] Use an accessor for rte_hash_key

Pablo, Also, what about the other patch? the use of a static variable in a recursive call? obviously incorrect for a threaded environment ... has that been accepted? regards, mts. On Thu, Oct 19, 2017 at 4:10 AM, De Lara Guarch, Pablo < pablo.de.lara.gua...@intel.com> wrote: > > > > -Orig

[dpdk-dev] What's the point of RTE_MBUF_PREFETCH_TO_FREE()?

It seems to be used to prefetch the mbuf in preparation for a free of an mbuf. But RTE_MBUF_PREFETCH_TO_FREE() currently does a prefetch for the ffirst cache line in the mbuf, but doesn't a free do a put to the mbuf's mempool? And isn't the mempool reference in the mbuf's second cache line? reg