On 1/15/2018 11:41 AM, Shahaf Shuler wrote: > Monday, January 15, 2018 1:34 PM, Ferruh Yigit: >>>> Currently no PMD seems implemented it. >>> >>> Sure, >>> >>> FAST_FREE offload is the logical AND between the old txqflags of: >>> ETH_TXQ_FLAGS_NOREFCOUNT >>> ETH_TXQ_FLAGS_NOMULTMEM >>> >>> The offload is just a performance optimization. As specified in the >> documentation [1] it enables the PMDs to further optimize the data path >> given the guarantees from the application. >>> Not having it means possible performance degradation for some PMD >> which rely on it. >>> >>> There is no PMD which implement it yet since not all PMDs moved to the >> new offloads API. However this flag is tested and translated into txqflags as >> part of rte_eth_convert_txq_offloads function. >>> Relevant PMDs for this offload will be: sfc, thunderx and i40e. >> >> Thank you for clarification, I am OK to have it. >> >> But since currently no PMD provide "DEV_TX_OFFLOAD_MBUF_FAST_FREE" >> capability, and default txq_flags is overwritten, some PMDs lost this >> optimization until they implement new capability, right? > > Yes this is correct, a good motivation to convert :).
My concern is being close to the integration deadline and if this patch cause performance drop for some PMDs. But I guess we can get the related PMD fixes almost until release if an issue observed, so this gives time to PMD maintainers. For series, Reviewed-by: Ferruh Yigit <ferruh.yi...@intel.com> > From the examples and testpmd part everything is ready. Just need for the PMD > to report back the capability to have this offload. >
