>>+ .ignore_offload_bitfield = 1, >>+ .offloads = DEV_RX_OFFLOAD_CRC_STRIP, > >It is not directly related to the patch. >May be I miss something, but it looks like there is no way to say that >"I always strip CRC and cannot preserve it".
Yes this is right. Not exposing the CRC offload flag means the device don’t support CRC strip toggling, however it does not explicitly say if device always strip/not. I guess device that has such limitation should specify it on the “Limitation” section of the PMD guide. >>+ txq_conf = dev_info.default_txconf; >>+ txq_conf.txq_flags = ETH_TXQ_FLAGS_IGNORE; >>+ txq_conf.offloads = port_conf.txmode.offloads; > >It looks like it is not 100% equivalent. As far as I can see dev_info get does >not convert txq_flags to offloads in default_txconf and in any case >txq_conf.offloads >are overwritten here. So, if PMD provides default txq_flags, it is lost. >If it is intentionally, it should be highlighted and explained. Yes it is. With the new Tx offloads API the application can choose the Tx offloads it wants to use according to its needs. For l2fwd case – it doesn’t use any of them. Any default txq flag the PMD set there is irrelevant. What I tried to do is not to preserve the entire old behavior rather to evolve the examples/applications while keeping the same functionality (i.e. the offloads which the application use are set, the rest are not). Moreover – it is a wrong approach, IMO, that the PMD set default offloads flags to the application. It has no knowledge to do so. I think this mechanism was initially created since the Tx offloads were all set by default, so it provided a mean to have good OOB configuration. Now when all offloads are set, I am not sure such API is needed anymore. Will be happy to hear more opinion on that. --Shahaf
