> -----Original Message----- > From: Wu, Jingjing > Sent: Thursday, January 22, 2015 7:38 AM > To: dev at dpdk.org > Cc: Wu, Jingjing; De Lara Guarch, Pablo; Cao, Min; Xu, HuilongX > Subject: [PATCH v2 0/6] new ntuple filter replaces 2tuple and 5tuple filters > > v2 changes: > - remove the code which is already applied in patch "Integrate ethertype > filter in igb/ixgbe driver to new API". > - modify commands' description in doc testpmd_funcs.rst. > > The patch set uses filter_ctrl API to replace old 2tuple and 5tuple filter > APIs. > It defines ntuple filter to combine 2tuple and 5tuple types. > It uses new functions and structure to replace old ones in igb/ixgbe driver, > new commands to replace old ones in testpmd, and removes the old APIs. > It removes the filter's index parameters from user interface, only the > filter's key and assigned queue are visible to user. > > Jingjing Wu (6): > ethdev: define ntuple filter type and its structure > ixgbe: ntuple filter functions replace old ones for 5tuple filter > e1000: ntuple filter functions replace old ones for 2tuple and 5tuple > filter > testpmd: new commands for ntuple filter > ethdev: remove old APIs and structures of 5tuple and 2tuple filters > doc: commands changed in testpmd_funcs for 2tuple amd 5tuple filter > > app/test-pmd/cmdline.c | 406 ++++++------- > app/test-pmd/config.c | 65 --- > doc/guides/testpmd_app_ug/testpmd_funcs.rst | 99 +--- > lib/librte_ether/rte_eth_ctrl.h | 57 ++ > lib/librte_ether/rte_ethdev.c | 116 ---- > lib/librte_ether/rte_ethdev.h | 192 ------ > lib/librte_pmd_e1000/e1000_ethdev.h | 69 ++- > lib/librte_pmd_e1000/igb_ethdev.c | 869 +++++++++++++++++++------ > --- > lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 468 +++++++++++---- > lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 52 +- > 10 files changed, 1300 insertions(+), 1093 deletions(-) > > -- > 1.9.3
Acked-by: Pablo de Lara <pablo.de.lara.guarch at intel.com> Just mind that the last patch (changing the documentation) does not apply properly, as there was another patch (from you I think), that modifies that document. Could you send another version of the last patch? Not sure if that's OK or if it is better to send the full patchset again.
