28/12/2017 22:54, Thomas Monjalon: > 22/12/2017 04:08, Ferruh Yigit: > > On 11/28/2017 2:09 PM, Thomas Monjalon wrote: > > > The pointer to the user parameter of the callback registration is > > > automatically pass to the callback function. > > > There is no point to allow changing this user parameter by a caller. > > > That's why this parameter is always set to NULL by PMDs and set only > > > in ethdev layer before calling the callback function. > > > > It can be helpful to mention a little from history: > > > > user cb_arg overwrite introduced by [1], to enable driver pass a pointer to > > user > > callback function and carry some information back from app to driver. > > > > Later to prevent usage of cb_arg as both input and output, an explicit > > output > > (ret_param) added [2] to _rte_eth_dev_callback_process() but previous cb_arg > > overwriting kept without any user. > > > > [1]: c1ceaf3ad056 ("ethdev: add an argument to internal callback function") > > > > [2]: d6af1a13d7a1 ("ethdev: add return values to callback process API") > > > > > > > > The NULL parameter in the internal callback processing function > > > is now removed. It makes clear that the callback parameter is user > > > managed and opaque from a DPDK point of view. > > > > > > Signed-off-by: Thomas Monjalon <tho...@monjalon.net> > > > > Reviewed-by: Ferruh Yigit <ferruh.yi...@intel.com> > > Applied
Sorry, not applied (should be applied on next-net). I will send a v2 with a more detailed explanation.