On Mon, Jan 2, 2017 at 6:02 PM, Sowmini Varadhan <sowmini.varad...@oracle.com> wrote: > On (01/02/17 17:31), Willem de Bruijn wrote: >> >> Thanks for adding this. >> >> walk_v3_tx is almost identical to walk_v1_v2_tx. That function can >> just be extended to add a v3 case where it already multiplexes between >> v1 and v2. > > I looked at that, but the sticky point is that v1/v2 sets up the > ring->rd* related variables based on frames (e.g., rd_num is tp_frame_nr) > whereas V3 sets these up based on blocks (e.g, rd_num is tp_block_nr) > so this impacts the core sending loop a bit.
Good point. Yes, deduplicating the function will help make it crystal clear where v3 differs from v2. The patch already has __v3_tx_kernel_ready and __v3_tx_user_ready, which can be plugged into the existing multiplexer functions __v1_v2_tx_kernel_ready and __v2_v2_tx_user_ready multiplexer (along with changing their names). We'll indeed need a similar multiplexer function for calculating the next frame to work around this rd_num issue, then.
